# Cast docs > Cast is a podcast finisher: it transcribes your recording so you can cut the audio by editing the text, cleans up filler words, pauses and noise, generates the intro, outro and background music the episode needs — with a commercial licence — and exports one finished file at the loudness your platform expects. # Getting started What Cast is, how your audio gets in, and how an episode goes from raw take to finished, licensed, publishable file. # What is Cast? Cast is a podcast finisher. Recording is the easy half — finishing is where episodes get stuck, so Cast handles the last mile: it transcribes what you recorded, lets you cut the audio by editing the text, cleans up the fillers and dead air, scores the episode with music that carries a commercial licence, and exports one finished file at the loudness your platform expects. **Where:** Runs in the browser. Nothing to install. ![A tour of Cast: a recording becomes a transcript, filler words and pauses are cleaned up, music is generated and ducks under the voice, and the finished episode is exported.](https://mubert.com/tools/cast/docs-media/showcase.mp4) This page has one job: orient you in thirty seconds — what Cast is, whether it is for you, and where to begin. The how-to for every feature lives on its own page. It is audio-first on purpose: no video timeline, no stock-music library to dig through — the music is generated for this episode, licensed, inside the same place you edit it. - **Solo hosts** — Record, clean, add music, export — without learning a DAW. - **Interview shows** — Tighten a conversation without losing its natural flow. - **Narrative podcasts** — Music and transitions make the story feel produced. - **Educators & experts** — Lectures and long talks become listenable episodes. ## Related - [Where do I start?](https://mubert.com/tools/cast/docs/where-to-start) - [How do I get my audio into Cast?](https://mubert.com/tools/cast/docs/import-audio) - [Which editor do I use — Voice or Main?](https://mubert.com/tools/cast/docs/which-editor) - [What do the plans include?](https://mubert.com/tools/cast/docs/plans-and-limits) --- Source: https://mubert.com/tools/cast/docs/what-is-mucast · Cast docs # Where do I start? On the Home screen. It walks you through your first episode in five steps — add audio, clean it up, edit by text, add music that ducks, export — and the New project button asks how you want to begin. You do not need to know the product to make the first episode; the product tells you. **Where:** Home — the first screen after you sign in. The five steps are not a tutorial to read; each one is a button that takes you to the right place with the right thing highlighted. Finish them once and you have shipped a real episode. ## Step by step: your first episode The Welcome project card tracks you through the whole path: add audio → clean it up → edit by text → music + ducking → export. Below it, Start something offers the same entrances as cards. ![The Home screen with the Welcome project card expanded: five numbered steps — Add audio, Clean it up, Edit by text, Music + ducking, Export episode — each a clickable stage of the first episode. Below, Start something cards: New voice project, Generate music, Record voice.](https://mubert.com/tools/cast/docs-media/home-start.png) ## Or just hit New project One question — how do you want to start? — and two live answers: with your voice, or with music. Video and templates are on their way. ![The New project dialog asking "How do you want to start?" with cards for Start with Voice (upload, transcribe, clean up) and Start with Music (generate or compose), plus Video and Template marked coming soon.](https://mubert.com/tools/cast/docs-media/new-project.png) ## Related - [What is Cast?](https://mubert.com/tools/cast/docs/what-is-mucast) - [How do I get my audio into Cast?](https://mubert.com/tools/cast/docs/import-audio) - [Which editor do I use — Voice or Main?](https://mubert.com/tools/cast/docs/which-editor) --- Source: https://mubert.com/tools/cast/docs/where-to-start · Cast docs # How do I get my audio into Cast? Upload a file or record straight in the browser — files up to 1 GB each on paid plans, 200 MB on Free. Either way transcription starts automatically as soon as the audio lands, so you can begin editing by text without asking for anything. **Where:** Upload page, Record page, or the Record chip on the voice lane in the Main Editor. **Plan:** How much you can bring in per month depends on your plan: 60 minutes on Free, 10 hours on Lite, 25 hours on Plus, unlimited on Max. Supported audio formats are MP3, WAV, M4A, AAC, OGG and FLAC — with one plan gate: on Free, lossless files (WAV and FLAC) are held for paid plans, so bring MP3 or M4A. Video files are rejected: Cast is audio-first on purpose — there is no video timeline to get lost in — so bring the audio track. That is what the Video category on the Upload page is for: the audio of a vlog or video, not the video itself. The per-file caps scale with the plan — 200 MB and 60 minutes per file on Free, 1 GB and 2 hours on paid — and on top of that sits the monthly allowance below. Browser recording captures one microphone at a time. Cast does not record several USB mics simultaneously — if you are recording two people, record them separately and upload both files. ## Upload a file Drop it, or bring a whole folder. WAV, MP3, M4A, AAC, FLAC (OGG works too), up to 1 GB per file on paid plans — and transcription starts on its own once it lands. ![The Upload page: a drop zone accepting WAV, MP3, M4A, AAC and FLAC up to the plan’s per-file limit, with uploads sorted into Voice, Music, SFX and Video categories.](https://mubert.com/tools/cast/docs-media/upload-page.png) ## Record in the browser One button, straight from your mic. Takes land on the Record page, then you choose where they go — or record onto the voice lane in the Main Editor. ![The Record page: one button captures a take straight from your microphone, and the takes land here so you can choose which project they go to.](https://mubert.com/tools/cast/docs-media/record-page.png) ## Related - [Which languages can Cast transcribe?](https://mubert.com/tools/cast/docs/transcription-languages) - [How do I edit the transcript — what can I actually do?](https://mubert.com/tools/cast/docs/edit-transcript) - [How do I separate two speakers in one audio track?](https://mubert.com/tools/cast/docs/separate-speakers) --- Source: https://mubert.com/tools/cast/docs/import-audio · Cast docs # Which languages can Cast transcribe? Cast auto-detects the language by default, and you can pin one of 15: English, Russian, Spanish, French, German, Italian, Portuguese, Japanese, Chinese, Korean, Arabic, Dutch, Polish, Turkish or Ukrainian. **Where:** Voice Editor — the language selector on the file. ![The language chip on a file reads AUTO LANG. Clicking it opens the picker — auto-detect plus fifteen languages. A language is pinned, then Re-transcribe is run from the file menu; the app warns that re-running ASR replaces the current transcript, and the words come back re-read.](https://mubert.com/tools/cast/docs-media/languages.mp4) Transcription runs automatically after an upload and costs no credits, on any plan, however many times you do it. Pin the language when auto-detect gets it wrong — most often on a short file, or one that opens with music before anyone speaks. Then re-transcribe from the file menu. Cast warns you first: re-running it replaces the current transcript and any in-line edits you made to it. ## Related - [How do I get my audio into Cast?](https://mubert.com/tools/cast/docs/import-audio) - [How do I fix a word the transcription got wrong?](https://mubert.com/tools/cast/docs/correct-transcript) - [What costs credits?](https://mubert.com/tools/cast/docs/credits-cost) --- Source: https://mubert.com/tools/cast/docs/transcription-languages · Cast docs # Is my work saved automatically? Yes. There is no Save button because there is nothing to press: every edit is saved to your account as you make it, survives a reload, and is there when you sign in from another computer. The one exception is a recording in progress: a take lives in the browser tab until you stop it, so closing the tab mid-take loses that take — see Can I pause a recording in progress? On top of the continuous save, Cast keeps version checkpoints of the whole project — see How does version history work? — so even an edit you made and later regret is recoverable. Exports are separate rendered files. Deleting or editing a project after exporting does not touch the files you already downloaded, and past exports stay available on the Exports page. ## Related - [How does version history work?](https://mubert.com/tools/cast/docs/version-history) - [Can I pause a recording in progress?](https://mubert.com/tools/cast/docs/can-i-pause-recording) - [Where do my past exports live?](https://mubert.com/tools/cast/docs/where-are-my-exports) --- Source: https://mubert.com/tools/cast/docs/does-cast-save · Cast docs # How does version history work? Cast checkpoints your whole project automatically — while you edit, and right before the risky moments: an export, Enhance voice, a re-transcribe, a bulk cleanup. Open the history from the clock icon in the top bar → "Version history…" and restore any checkpoint. **Where:** Either editor → clock icon in the top bar → Version history… You never create a checkpoint by hand and you cannot forget to. While you are actively editing, Cast cuts one roughly every half hour of work; on top of that it always takes one just before anything that changes a lot at once — so "Before export" and "Before re-transcribe" are sitting there when you need them. Each entry in the list says what it was, not a version number. Restoring is safe by construction: before a restore, Cast checkpoints the state you are leaving, and the restore itself becomes a new step in the history. Nothing is ever deleted from the timeline — if you restore and change your mind, restore forward again. If you would rather compare than roll back, Duplicate turns any checkpoint into a new project and leaves the current one exactly as it is. On the free plan you see the last 7 days of checkpoints. Older ones are kept, not thrown away — upgrading unlocks the full history of every project retroactively. ## Related - [Is my work saved automatically?](https://mubert.com/tools/cast/docs/does-cast-save) - [Can I undo a cleanup — even long after I made it?](https://mubert.com/tools/cast/docs/undo-cleanup) --- Source: https://mubert.com/tools/cast/docs/version-history · Cast docs # Which editor do I use — Voice or Main? Use the Voice Editor for anything about what was said: filler words, pauses, wording, speakers, chapters. Use the Main Editor for anything about how the episode sounds: music, sound effects, ducking, effects, export. **Where:** Both open from a project — the tabs at the top left switch between them. The Voice Editor shows you words, not a waveform. The Main Editor shows you a timeline of clips and lanes. Your edits carry over. Clean up the speech first, then move to the mix — every cut you made is already applied to the voice clip you find there. ## Voice — what was said The transcript is the interface: filler words, pauses, wording, speakers, chapters. The audio sits above as a thin strip you rarely touch. ![The same project on the Voice tab (underlined, top left): the transcript fills the screen — words with strikethrough cuts, speakers, chapters — and the audio is only a thin strip above. This editor is about what was said.](https://mubert.com/tools/cast/docs-media/which-editor-voice.png) ## Main — how it sounds The same project, one tab over: lanes and clips. Voice on one lane, music ducking under it, sound effects and export — the mix. ![The same project on the Main tab (underlined, top left): a lane-based timeline — the voice waveform on one lane, music with an auto-ducking badge on another, SFX below. This editor is about how the episode sounds.](https://mubert.com/tools/cast/docs-media/which-editor-main.png) ## Related - [How do I edit the transcript — what can I actually do?](https://mubert.com/tools/cast/docs/edit-transcript) - [How does the timeline work?](https://mubert.com/tools/cast/docs/timeline-clips) - [Why does the music get quieter when someone talks?](https://mubert.com/tools/cast/docs/auto-ducking) --- Source: https://mubert.com/tools/cast/docs/which-editor · Cast docs # Editing the transcript Edit the audio by editing the text: cut words, remove fillers, trim pauses, label speakers. # How do I edit the transcript — what can I actually do? Select any run of words in the Voice Editor and a menu appears. You can delete them (the audio is cut and the gap closes), ignore them (the audio is muted, the timing stays), correct the wording, mark a chapter, assign a speaker, or copy and paste words elsewhere. **Where:** Voice Editor — select words in the transcript. Deleting words is a real edit to the audio, not just to the text. The recording is cut and the timeline gets shorter, and because everything is non-destructive, it can be restored later. Press Enter on a word to split the paragraph there — that is how you break up a wall of text into readable turns. Find & Replace works across the whole transcript, which is the fastest fix when a name or a term is mis-heard the same way throughout. And a toolbar toggle lets you show or hide the text you have deleted, so you can see what you cut. ## Delete — cut it out of the audio The words vanish, the audio underneath is ripple-cut, and everything after slides earlier — watch the episode duration drop. Reversible from Recent edits, like every cut. ![A run of words is selected and Delete is pressed. The words vanish and the episode duration drops — the audio was cut, not just the text.](https://mubert.com/tools/cast/docs-media/action-delete.mp4) ## Ignore — mute it, keep the timing The words are struck out and silenced, but the gap keeps its length, so nothing after it moves. Use it when the audio is cut to video, or a music bed underneath must not drift. ## Correct — fix the text, not the voice Type over a mis-heard word or phrase; transcript, captions and exports update. The recording is untouched — Cast does not synthesize words into your voice. Double-click any single word to edit it in place. ![A mis-heard phrase 'the I native products' is selected and rewritten to 'AI-native products' in the popover, then a single word is double-clicked and 'Lilia' is fixed to 'Lillia'. The transcript, captions and exports update; the waveform above never changes — the recording is untouched.](https://mubert.com/tools/cast/docs-media/correct.mp4) ## Keep — clear a filler flag When the filler detector flags a word you actually want — a deliberate "like", a real "so" — Keep clears the mark so bulk cleanup leaves it alone. ## Chapter — two ways, rename, fold Make a chapter from a selection (⌘⇧M, titled from your words) or click "+ Chapter here" between paragraphs — no selection needed. Double-click the title to rename it (a single click seeks there); fold the section to work with the structure instead of the wall of text. The same chapters ride the Main timeline. ![A chapter is made two ways — from a selected sentence, and by clicking '+ Chapter here' between paragraphs with nothing selected. The default 'Chapter 1' is renamed by double-clicking its title, and the whole section folds away by clicking the thin bar on its left.](https://mubert.com/tools/cast/docs-media/chapters.mp4) ## Speaker — label who is talking Assign the selection from the menu, or press ⇧S for the same picker — or just press the speaker’s digit and skip the picker entirely. Speakers are saved to your library and reusable across projects. ![A line is reassigned to a different speaker three ways: the Speaker button in the selection menu opens a picker, ⇧S opens the same picker without the mouse, and pressing a digit assigns the speaker who owns that number with no picker at all. The turn re-labels and recolours each time.](https://mubert.com/tools/cast/docs-media/speakers.mp4) ## Related - [What is the difference between Delete and Ignore?](https://mubert.com/tools/cast/docs/delete-vs-ignore) - [How do I fix a word the transcription got wrong?](https://mubert.com/tools/cast/docs/correct-transcript) - [Can I undo a cleanup — even long after I made it?](https://mubert.com/tools/cast/docs/undo-cleanup) - [How do I remove filler words?](https://mubert.com/tools/cast/docs/remove-filler-words) - [Bleep swearing and sensitive words — the whole episode, one pass](https://mubert.com/tools/cast/docs/censor-swearing) --- Source: https://mubert.com/tools/cast/docs/edit-transcript · Cast docs # How do I remove filler words? The Voice Editor finds them for you and lists them. Hesitations — "um", "uh", "mm" — are flagged by default and you can cut them all in one click. Discourse markers like "like" and "so" are detected too, but stay off until you switch them on word by word. **Where:** Voice Editor → Fillers. One click, and no chopped words. Every filler cut is snapped to a quiet point in the waveform, so the edit lands between the sounds instead of through them — and when Cast cannot find clean edges, it mutes the word rather than leave you a click. That is the difference between "cut all the umms" being a button you trust and one you undo. Leaving "like" and "so" off by default is deliberate. They are real words as often as they are filler, and flagging every one of them would bury the transcript in false positives. You decide which ones count, by clicking them in the dictionary. You can add your own words, and the dictionary can differ per speaker — which is what makes cleanup practical on a two-host show where each person has their own tics. ## The panel: every hit, counted Cast lists each one against the transcript it came from, grouped by chapter and counted per speaker. Cut them one at a time, or take the lot in a single click — and Auto-advance walks you to the next one so you are not hunting. The chips above the list — Hesitations, Discourse, Custom — are filters, not just counters. Click one to hide or show that kind of filler, and the bulk buttons follow what is visible: narrow the list to hesitations only and "Cut all" cuts exactly those. ![The Fillers panel beside the transcript: 36 hits found, split into Hesitations and Discourse and counted per speaker, each listed with its timecode and a Cut button, plus Cut all for the lot.](https://mubert.com/tools/cast/docs-media/fillers-panel.png) ## The dictionary: what counts as a filler is your call Hesitations are on: "uh", "um", "uhm", "er", "hmm" — noises, not words. Discourse markers are off, and struck through until you click them: "i mean", "basically", "literally", "sort of", "right", "so". They are real words as often as they are filler, and flagging them all would bury the transcript in false positives. Click a word to flip it. Add your own. Keep a different list per speaker — a two-host show where each person has their own tics is exactly where a single global list falls apart. ![The filler dictionary expanded: HESITATIONS (uh, um, uhm, umm, er, erm, ah, eh, hmm, mm, mmm) all active; DISCOURSE MARKERS, off by default, with "like" and "you know" switched on and "i mean", "basically", "literally", "sort of", "kind of", "right", "so", "well", "actually", "honestly" struck through — not flagged.](https://mubert.com/tools/cast/docs-media/fillers-dictionary.png) ## Related - [Why isn’t every "like" flagged as a filler?](https://mubert.com/tools/cast/docs/fillers-not-all-flagged) - [What exactly does Cut all cut — and can I cut just some of them?](https://mubert.com/tools/cast/docs/cut-all-fillers) - [How do I remove pauses and silences?](https://mubert.com/tools/cast/docs/remove-pauses) - [Can I remove mouth clicks and lip smacks?](https://mubert.com/tools/cast/docs/mouth-clicks) - [What is the difference between Delete and Ignore?](https://mubert.com/tools/cast/docs/delete-vs-ignore) - [How do I separate two speakers in one audio track?](https://mubert.com/tools/cast/docs/separate-speakers) - [Bleep swearing and sensitive words — the whole episode, one pass](https://mubert.com/tools/cast/docs/censor-swearing) --- Source: https://mubert.com/tools/cast/docs/remove-filler-words · Cast docs # Why isn’t every "like" flagged as a filler? Because "like" is off until you switch it on. Hesitations — "um", "uh", "mm" — are flagged everywhere by default; real words that double as filler, like "like", "so" and "you know", wait in the dictionary until you enable them. One click there flags every instance in the transcript. **Where:** Voice Editor → Fillers → Dictionary. The dictionary sorts fillers into two kinds, and they behave differently on purpose. Hesitations are noises — "uh", "um", "erm", "hmm" — that carry no meaning, so Cast flags every one of them out of the box. Discourse markers are real words: "like", "so", "right", "actually", "you know". A transcript where every "so" is flagged is a transcript you stop trusting, so those stay off until you say otherwise. Enabling a word is all-or-nothing in the good sense: click "like" in the dictionary and every "like" in the episode is flagged — the whole transcript is re-scanned on the spot, not just the spots the initial analysis happened to notice. The Fillers count updates immediately, and so does the transcript highlighting. The dictionary matches the word, not the grammar. "I like your show" gets flagged along with "it was, like, weird" — no detector can hear the difference reliably, so Cast does not pretend to. The escape hatch is per-instance: click a flagged word you want to keep and choose Keep. That clears the one flag without touching the dictionary. Your dictionary travels with the project. Enable "like", add your own words, strike out a built-in — reopen the project on another machine and the same rules apply. Each speaker can also carry their own word list, so one host’s "basically" habit does not flag the other’s vocabulary. - **Hesitations** — Noises, not words — "uh", "um", "hmm". Flagged by default, everywhere. - **Discourse markers** — Real words that double as filler — "like", "so", "you know". Off (struck through) until you click them on. - **Custom words** — Anything you add yourself — always flagged. Adding the word is the opt-in. - **Keep** — Clears the flag on one instance without changing the dictionary. ## Related - [How do I remove filler words?](https://mubert.com/tools/cast/docs/remove-filler-words) - [What exactly does Cut all cut — and can I cut just some of them?](https://mubert.com/tools/cast/docs/cut-all-fillers) - [How do I separate two speakers in one audio track?](https://mubert.com/tools/cast/docs/separate-speakers) --- Source: https://mubert.com/tools/cast/docs/fillers-not-all-flagged · Cast docs # What exactly does Cut all cut — and can I cut just some of them? Cut all cuts what the list is currently showing — so the filter chips are your scalpel. Narrow the list to Hesitations only, or to one speaker, and the button follows: "Cut all (21)" means those 21. Every cut is snapped to a quiet point, the lot lands as one undo step, and a checkpoint is saved first. **Where:** Voice Editor → Fillers → Cut all. The button is honest about its number. "Cut all (142)" counts what is in the list right now: filters applied, and anything already cut or ignored excluded. Narrow the list and the number follows. Two sets of chips do the narrowing. The category chips — Hesitations, Discourse, Custom — toggle each kind in and out of the list. The speaker chips narrow to one voice, and the button says so: "Cut all (36) · Maria" cuts Maria’s fillers and nobody else’s. A practical recipe: cut the umms blind, review the likes by hand. Toggle off Discourse and Custom, hit Cut all — hesitations are noise, nothing of value is lost. Then toggle them back on and walk the remaining hits one by one with Auto-advance, keeping the "like"s that are doing real work in the sentence. Bulk does not mean blunt. Each filler in the batch is cut individually: the edit is snapped to a quiet point in the waveform so it lands between sounds, and where no clean edge exists, that word is muted instead of clipped. A hundred fillers cut at once are a hundred separately-judged edits. And it is dressed in seatbelts. The whole batch is one undo step (⌘Z takes it all back), a project checkpoint is saved automatically right before the cut, and every individual filler stays restorable afterwards — from Recent edits, or from its "✂" mark in the transcript. - **Category chips** — Hesitations / Discourse / Custom — toggle kinds in and out; Cut all follows the visible set. - **Speaker chips** — Narrow to one speaker; the button names who you are about to cut. - **Mute all** — The same batch, muted instead of cut — timing stays put. See Delete vs Ignore. ## Related - [How do I remove filler words?](https://mubert.com/tools/cast/docs/remove-filler-words) - [Why isn’t every "like" flagged as a filler?](https://mubert.com/tools/cast/docs/fillers-not-all-flagged) - [What is the difference between Delete and Ignore?](https://mubert.com/tools/cast/docs/delete-vs-ignore) - [Can I undo a cleanup — even long after I made it?](https://mubert.com/tools/cast/docs/undo-cleanup) - [I broke my project — how do I get an earlier version back?](https://mubert.com/tools/cast/docs/restore-version) --- Source: https://mubert.com/tools/cast/docs/cut-all-fillers · Cast docs # How do I remove pauses and silences? Cast lists every silence and marks it in the transcript. You set the threshold — show me pauses longer than X seconds, 0.5s by default — and then cut or shorten them, one by one or all at once. **Where:** Voice Editor → Pauses. ![The threshold in the Pauses panel is dragged from 0.8 to 1.8 seconds. The list of silences re-filters instantly as it moves and the count drops — the control is a real number of seconds, not an abstract sensitivity, and no re-analysis is needed.](https://mubert.com/tools/cast/docs-media/remove-pauses.mp4) The control is an explicit length in seconds, not an abstract sensitivity slider, so you always know exactly which pauses you are about to act on. Changing it re-filters instantly. Shorten is usually the right action. A pause that is merely too long wants trimming to a beat — 0.3 seconds by default, and the Shorten to slider sets any length up to a second — while cutting it out entirely makes speech sound unnaturally jammed together. - **Cut** — Removes the silence, closes the gap. - **Shorten** — Trims it to a length you set — 0.3s by default. ## Related - [How do I remove filler words?](https://mubert.com/tools/cast/docs/remove-filler-words) - [What is the difference between Delete and Ignore?](https://mubert.com/tools/cast/docs/delete-vs-ignore) --- Source: https://mubert.com/tools/cast/docs/remove-pauses · Cast docs # Can I remove mouth clicks and lip smacks? Yes. Mouth sounds between words are detected acoustically and flagged by default. Because the detection listens to the sound rather than matching a word list, it works in any language. **Where:** Voice Editor → Mouth. ![The Mouth sounds panel beside the transcript: 98 clicks, smacks and breaths found, each listed with its timecode and a Cut button, and a real Sensitivity slider — turn it up to catch more, down to keep the breaths you want.](https://mubert.com/tools/cast/docs-media/mouth-panel.png) This one gets a real sensitivity slider — unlike filler words, acoustic confidence has a genuine gradient. Turn it up to catch more, down if it is flagging breaths you would rather keep. Cut is the right action here. Muting a click leaves a small hole where the sound was, which is usually more noticeable than the click. ## Related - [How do I remove filler words?](https://mubert.com/tools/cast/docs/remove-filler-words) - [How do I remove background noise?](https://mubert.com/tools/cast/docs/remove-background-noise) --- Source: https://mubert.com/tools/cast/docs/mouth-clicks · Cast docs # How do I fix a word the transcription got wrong? Select the word and choose Correct, then type what it should say. This fixes the transcript, the captions and the exported text — it does not change the recorded audio, which still says what was actually said. **Where:** Voice Editor → select a word → Correct. ![A mis-heard phrase 'the I native products' is selected and rewritten to 'AI-native products' in the popover, then a single word is double-clicked and 'Lilia' is fixed to 'Lillia'. The transcript, captions and exports update; the waveform above never changes — the recording is untouched.](https://mubert.com/tools/cast/docs-media/correct.mp4) That distinction matters and it is worth stating plainly: correcting text is not re-recording. Cast does not synthesize a new word into your voice. If you want the audio itself to say something different, you have to record it. For a name or a term the transcription gets wrong the same way every time, use Find & Replace instead of fixing each one. ## Related - [How do I edit the transcript — what can I actually do?](https://mubert.com/tools/cast/docs/edit-transcript) - [How do I export captions or subtitles?](https://mubert.com/tools/cast/docs/captions-subtitles) --- Source: https://mubert.com/tools/cast/docs/correct-transcript · Cast docs # What is the difference between Delete and Ignore? Delete cuts the audio out and closes the gap, so the episode gets shorter. Ignore mutes the audio but leaves the gap, so nothing after it moves. Both are reversible, and both remove the words from captions and exports. **Where:** Voice Editor — both are offered on any selection, filler or mouth click. (Pauses get Cut and Shorten — muting a gap would change nothing.) Delete is the right default for speech junk. An ignored filler word leaves a small silent hole that breaks the rhythm of the sentence, which usually sounds worse than the filler did. Ignore is right when timing must not move — when the audio is cut to video, or a music bed underneath would drift out of place if the voice got shorter. ## Related - [Can I undo a cleanup — even long after I made it?](https://mubert.com/tools/cast/docs/undo-cleanup) - [How do I remove pauses and silences?](https://mubert.com/tools/cast/docs/remove-pauses) - [How do I edit the transcript — what can I actually do?](https://mubert.com/tools/cast/docs/edit-transcript) - [Bleep swearing and sensitive words — the whole episode, one pass](https://mubert.com/tools/cast/docs/censor-swearing) --- Source: https://mubert.com/tools/cast/docs/delete-vs-ignore · Cast docs # Can I undo a cleanup — even long after I made it? Yes. Undo and redo work as normal, and separately, every cut and mute is listed under Recent edits with its own Restore. Because Cast never overwrites your recording, you can put back one filler word you removed twenty edits ago without losing anything you did since. **Where:** Voice Editor → Recent edits. This is the practical payoff of a non-destructive model. In tools that bake AI cleanup into the audio file, an over-aggressive pass is a problem you solve by starting over. Here it is a problem you solve by clicking Restore. The same is true after an export: the export is a rendered copy, and your project still holds the original recording plus your edits. ## Related - [What is the difference between Delete and Ignore?](https://mubert.com/tools/cast/docs/delete-vs-ignore) - [How do I edit the transcript — what can I actually do?](https://mubert.com/tools/cast/docs/edit-transcript) --- Source: https://mubert.com/tools/cast/docs/undo-cleanup · Cast docs # How do I separate two speakers in one audio track? Cast does not auto-detect who is speaking. You label speakers by hand: select the text and assign a speaker (⇧S), or set the speaker for a whole paragraph. Speakers are saved to your library and reusable across projects, and each one can have its own filler-word dictionary. **Where:** Voice Editor → Speakers, or select text → Speaker. ![A line is reassigned to a different speaker three ways: the Speaker button in the selection menu opens a picker, ⇧S opens the same picker without the mouse, and pressing a digit assigns the speaker who owns that number with no picker at all. The turn re-labels and recolours each time.](https://mubert.com/tools/cast/docs-media/speakers.mp4) Be clear about what this does and does not do. Labelling tells the transcript who said what. It does not split one mixed recording into separate per-person audio tracks, and it does not remove mic bleed — no editor does that reliably from a single mixed file, whatever the marketing says. If you have not recorded yet, the fix is upstream: give each person their own microphone and upload the files separately. That is the only way to get truly separate speaker audio, and it is what every tool in this category will tell you. Press Enter on a word to split a paragraph, so a turn that runs across a hand-off can be divided and assigned correctly. ## Related - [How do I remove filler words?](https://mubert.com/tools/cast/docs/remove-filler-words) - [How do I get my audio into Cast?](https://mubert.com/tools/cast/docs/import-audio) - [How do I add chapters with timestamps?](https://mubert.com/tools/cast/docs/chapters) --- Source: https://mubert.com/tools/cast/docs/separate-speakers · Cast docs # Can I listen faster without the chipmunk voice? Yes. The speed control in the transport runs from 0.5× to 2× and preserves the pitch, so a 2× pass through an episode still sounds like a person — which is how you proof an hour of audio in half an hour. **Where:** The transport bar in either editor — the ×-button next to play. Speed changes playback only. It never touches the recording or the export, so leaving it on 2× costs nothing. The practical use is review: skim the episode fast, and drop back to 1× when something needs a careful ear. ## Related - [How do I edit the transcript — what can I actually do?](https://mubert.com/tools/cast/docs/edit-transcript) - [Can I undo a cleanup — even long after I made it?](https://mubert.com/tools/cast/docs/undo-cleanup) --- Source: https://mubert.com/tools/cast/docs/playback-speed · Cast docs # How do I add chapters with timestamps? Select the text where a chapter should start and mark it (⌘⇧M) — the title comes from your selection. Chapters export as timestamp lists for YouTube and for Podcasting 2.0. **Where:** Voice Editor, and the chapters layer in the Main Editor. ![A chapter is made two ways — from a selected sentence, and by clicking '+ Chapter here' between paragraphs with nothing selected. The default 'Chapter 1' is renamed by double-clicking its title, and the whole section folds away by clicking the thin bar on its left.](https://mubert.com/tools/cast/docs-media/chapters.mp4) Chapters are anchored to the words you attached them to, not to a clock time. That is what keeps them right after you cut two minutes of filler out of the middle: a timestamped chapter would now point at the wrong moment, while yours still points at the sentence you chose. ## Related - [How do I export chapters with timestamps?](https://mubert.com/tools/cast/docs/export-chapters) - [How do I separate two speakers in one audio track?](https://mubert.com/tools/cast/docs/separate-speakers) --- Source: https://mubert.com/tools/cast/docs/chapters · Cast docs # Bleeping & censoring Bleep swearing, redact a name or a spoiler, and publish a clean version that stays clean in the audio and the captions — a dictionary scan for the swears, a manual bleep for the rest, all non-destructive. # Bleep swearing and sensitive words — the whole episode, one pass The Voice Editor scans your transcript against a profanity dictionary in the episode’s language and lists every hit. Censor them all in one click or one at a time, and pick the mask — a beep, a noise, or silence. The word is masked in the transcript and the captions too, so a clean version is clean everywhere. **Where:** Voice editor → Censor (right-hand rail). This is the fast way to make a clean cut of an episode — the version a brand-safe sponsor asks for, or the one you publish without the explicit tag. Doing it by hand means finding each word by ear and dropping a beep over it; here the words are already on the page, so Cast finds them and you approve them. Censoring is non-destructive, like every edit in Cast. A censored word is muted under the mask, not erased, so you can restore it later — and switch a beep to silence, or a bleep to brown noise, without redoing the work. One honest limit: a bleeped episode is not automatically brand-safe or advertiser-approved. Marking a show clean or explicit is your call, and platforms leave the judgement to you. Cast gives you a fast, reversible way to produce the clean version — not a guarantee about how a platform or sponsor will treat it. ## Set it up: the sound, the mask style, the dictionaries One screen sets the pass: the cover sound for the episode (Beep 1 kHz by default), Show as — how a censored word reads in the transcript and captions (f***, ****, or [bleep]) — and which dictionaries flag words: the built-in Strong-profanity list for the episode’s language, plus My words for your own terms. Hit Go and Cast scans the whole transcript. ![The Censor setup panel: episode sound (Beep 1 kHz with a change dropdown), a Show as switch for the mask style (f***, ****, [bleep]), the Strong-profanity dictionary of 279 words, a My words box with one custom term, and a Go button.](https://mubert.com/tools/cast/docs-media/censor-setup-panel.png) ## Scan the whole episode, bleep the lot in one click Cast lists every flagged word with its timecode — the word itself in bold, the line it came from alongside — so you can judge each hit without hunting for it. Click a row and the transcript jumps there, with the word highlighted and the playhead parked on it. Censor the lot with Censor all, or take them one at a time; the ⊘ button tells the scan to never flag that word again. ![The Censor results list: “1 found · 44 censored”, the flagged word in bold with its surrounding line and timecode, a Censor action on the row, and a Censor all (1) button below.](https://mubert.com/tools/cast/docs-media/censor-found-list.png) ## Everything you censored, in one list Under the scan results sits Censored (N) — the running list of every bleep in the episode, styled like the Pauses and Fillers lists: the timecode, the covered word, and the cover it plays (Beep 1 kHz, Silence, …). Click a row to see that word in the transcript; Restore on the row brings just that one back, Restore all clears the lot. ![The Censored (56) change-list in the Censor panel: each row shows a timecode, the censored word, and its cover sound (Silence), with a Restore action per row and a “show in the transcript” tooltip.](https://mubert.com/tools/cast/docs-media/censor-change-list.png) ## Beep, noise, or silence — the cover is yours, per word The cover over a censored word is your choice: three beep tones, white, brown or static noise, a crackle, or plain silence. Set a default for the episode, and override it on any single word — click a bleeped word in the transcript and its popover flips between Silence and Beep, or opens the full sound library, each entry previewable in place. ![A censored word’s popover in the transcript — play, a Silence/Beep toggle, the current sound, and Restore — with the sound library open above: Silence, three beep tones (1 kHz, 800 Hz, 1.2 kHz), brown and white noise, and static cracks and noise, each with a preview button.](https://mubert.com/tools/cast/docs-media/censor-popover-sounds.png) ## Add your own words — and protect the ones you meant to keep The built-in dictionary covers profanity in the episode’s language. Add your own terms — a client’s name, a spoiler, an unreleased product — and they are flagged the same way; each My-words term can also carry its own cover sound instead of the episode default. And when the dictionary catches a word you meant to keep, mark it Never and it stays untouched, this episode and the next. ![The Censor setup: the Strong-profanity dictionary of 279 words (struck-through words are the ones kept unflagged), with a highlighted My words box for adding your own terms.](https://mubert.com/tools/cast/docs-media/censor-dictionary-mywords.png) ## Select and bleep anything — names, spoilers, brands Not every word you want to cover is a swear. Select any phrase in the transcript — a client’s name, an unreleased product, a spoiler you talked around — and the selection popover has a Censor action that drops a bleep over exactly those words. The text stays on the page; the audio underneath is masked. It is the manual counterpart to the dictionary scan, for the one-offs a word list can’t know about. To change the cover on a word already bleeped, click it: the popover swaps its beep for noise or silence, and offers Apply to all when the same word appears more than once — so every instance changes together. ![A phrase selected in the transcript with the actions popover above it — Speaker, Correct, Ignore, Censor, Delete, Copy, Chapter — where the Censor action drops a bleep over exactly the selected words.](https://mubert.com/tools/cast/docs-media/censor-manual-select.png) ## How the word reads on the page: masked, readable, or dimmed The text side of a bleep is a display choice. Masked hides the word behind the mask style you picked — f***, ****, or [bleep] — Readable keeps it legible for your own editing, and Dimmed greys it out without hiding it. The mask style is one setting surfaced in three places — the Censor panel’s Show as row, the transcript’s Aa Display menu, and the Export dialog — change it in any of them and the transcript view and every text export follow. ![The transcript’s Aa Display menu with the Censored words section: Masked, Readable, and Dimmed display modes, and the mask style choices f***, ****, and [bleep].](https://mubert.com/tools/cast/docs-media/censor-display-modes.png) ## The bleep ships in the file and the captions, not just on screen A censored word is masked everywhere the episode goes, not only in the editor. On export, Cast hard-mutes the original word in the rendered audio and lays the beep over it — the mask is baked into the MP3, not a screen preview — so the file you publish is genuinely clean. The captions are masked to match, so the word never leaks in the transcript a platform reads. Audio and text stay clean together. ## Related - [How do I remove filler words?](https://mubert.com/tools/cast/docs/remove-filler-words) - [What is the difference between Delete and Ignore?](https://mubert.com/tools/cast/docs/delete-vs-ignore) - [How do I export captions or subtitles?](https://mubert.com/tools/cast/docs/captions-subtitles) - [Can I undo a cleanup — even long after I made it?](https://mubert.com/tools/cast/docs/undo-cleanup) --- Source: https://mubert.com/tools/cast/docs/censor-swearing · Cast docs # Does bleeping change my episode’s length? No. A bleep sits on top of the word — the original is muted for that span and the cover plays over it — so the episode runs exactly as long as before. Deleting a word is the one that shortens the timeline; censoring only covers. **Where:** Voice editor → Censor (right-hand rail). Censoring is a cover, not a cut. The word underneath is muted and a beep, noise, or silence plays over that exact span; nothing is removed, so the runtime, your chapters, and every music cue all stay where they were. If you would rather the word were gone than covered, that is a Delete instead — select it in the transcript and delete, and the audio is cut and the episode gets shorter. A simple rule: Censor to cover a word, Delete to remove it. ## Related - [Bleep swearing and sensitive words — the whole episode, one pass](https://mubert.com/tools/cast/docs/censor-swearing) - [What is the difference between Delete and Ignore?](https://mubert.com/tools/cast/docs/delete-vs-ignore) - [How do I edit the transcript — what can I actually do?](https://mubert.com/tools/cast/docs/edit-transcript) --- Source: https://mubert.com/tools/cast/docs/censor-vs-delete-length · Cast docs # Can I undo a bleep or change the cover later? Yes — every bleep is non-destructive. Restore a single word from its popover, or Restore all to clear them at once, and the original audio comes back untouched. You can also swap a beep for noise or silence at any time without redoing the work. **Where:** Voice editor → Censor (right-hand rail). ![The Censored (56) list in the Censor panel: every bleeped word with its timecode and cover sound, a Restore action on each row, and a Restore all button above the list.](https://mubert.com/tools/cast/docs-media/censor-change-list.png) A censored word is muted under the cover, never erased, so nothing you bleep is permanent. Click a bleeped word and choose Restore to bring it back, restore one row from the Censored list in the panel, or use Restore all to clear the lot — the audio returns exactly as recorded. Changing your mind about the cover is just as cheap: the same popover swaps a hard beep for brown noise or plain silence, and Apply to all pushes that choice to every other instance of the word in one move. ## Related - [Bleep swearing and sensitive words — the whole episode, one pass](https://mubert.com/tools/cast/docs/censor-swearing) - [Can I undo a cleanup — even long after I made it?](https://mubert.com/tools/cast/docs/undo-cleanup) --- Source: https://mubert.com/tools/cast/docs/censor-undo-restore · Cast docs # Does bleeping make my episode brand-safe or advertiser-approved? Not by itself. Bleeping is a fast, reversible way to produce a clean version — the one a sponsor asks for or the episode you publish without the explicit tag — but marking a show clean or explicit, and whether a platform or advertiser accepts it, is a judgement they make, not something Cast certifies. **Where:** Voice editor → Censor (right-hand rail). What Cast gives you is the clean cut itself: the flagged words covered in the audio and masked in the captions, ready to export. That is the practical thing sponsors and directories ask for. What it cannot do is promise how anyone will treat the result. Whether an episode counts as clean, and whether a sponsor or platform signs off, is their call — so bleep for the clean version you want to publish, not as a guarantee about someone else’s policy. ## Related - [Bleep swearing and sensitive words — the whole episode, one pass](https://mubert.com/tools/cast/docs/censor-swearing) --- Source: https://mubert.com/tools/cast/docs/censor-brand-safe · Cast docs # What languages does the profanity dictionary cover? The built-in Strong-profanity list follows your episode’s detected language, across 25 languages — English, Spanish, French, German, Portuguese, Italian, Dutch, Russian, Polish, Japanese, Korean, Chinese, Arabic, Hindi and more. Anything it doesn’t catch — or a language it doesn’t cover — you add to My words. **Where:** Voice editor → Censor (right-hand rail). Cast detects the language of each upload and scans it against the matching list, so a Spanish episode is checked against Spanish profanity, not English. The dictionaries cover 25 languages between them. The built-in lists are deliberately conservative — they catch the obvious words, not every borderline one. Whatever they miss, and any term specific to your show — a name, a spoiler, a product still under wraps — goes in My words and is flagged exactly the same way. ## Related - [Bleep swearing and sensitive words — the whole episode, one pass](https://mubert.com/tools/cast/docs/censor-swearing) --- Source: https://mubert.com/tools/cast/docs/censor-languages · Cast docs # Improving the sound Noise, room tone and uneven levels — fixed without making you sound like a robot, and reversible if it does. # How do I remove background noise? Turn on the AI Noise Remover. It strips hiss, hum and room noise from the voice while leaving music and sound effects alone, and it has four settings — Off, Subtle, Medium, Strong — so you can control how hard it works. **Where:** Voice Editor → Polish, or the Polish step when audio comes in. ![The Polish panel beside the transcript: AI Noise Remover, described as a neural denoise that removes hiss, hum and room noise while preserving music and SFX, with four settings — Off, Subtle, Medium, Strong.](https://mubert.com/tools/cast/docs-media/noise-remover.png) Start at Subtle and go up only if you still hear the noise. The instinct is to reach for Strong immediately; resist it. Heavy noise reduction is what makes a voice sound processed and hollow, and once you have heard that in your own recording you cannot un-hear it. It takes seconds per file, and you can keep editing while it runs. It is one of two polish modes — the other, Enhance voice, is the deep pass — and only one can be active at a time, because they are the same cleanup at different depths. Noise removal costs no credits, on any plan. ## Related - [How do I make my voice sound more "studio"?](https://mubert.com/tools/cast/docs/speech-enhancer) - [The AI cleanup made my voice sound robotic — how do I fix it?](https://mubert.com/tools/cast/docs/sounds-robotic) - [Can I remove echo or room reverb from a recording?](https://mubert.com/tools/cast/docs/remove-echo) --- Source: https://mubert.com/tools/cast/docs/remove-background-noise · Cast docs # How do I make my voice sound more "studio"? Turn on Enhance voice. It separates your voice from the background, removes the noise and brings it to a consistent studio loudness — at Subtle, Medium or Strong. Budget roughly two minutes of processing per ten minutes of audio. **Where:** Voice Editor → Polish (the second row). Also on a voice clip in the Main Editor, and as the Enhance voice toggle in Export, which runs it over every voice clip at once. ![The Enhance voice section in the clip inspector, switched off: a single toggle described as "Removes background noise, room echo, and matches volume". Turning it on reveals the three strengths — Subtle, Medium, Strong.](https://mubert.com/tools/cast/docs-media/enhance-voice.png) The difference from plain noise removal: the Noise Remover is fast and only takes noise away, while this rebuilds the sound — it lifts the voice away from the background first, cleans it, then fixes the level, so a voice that drifts between loud and quiet comes back even. That levelling is usually what people actually mean when they say a recording sounds amateur. It is the slower of the two for that reason: seconds, versus about two minutes for every ten minutes of audio. You can run one polish at a time, not both. That is deliberate: Enhance voice already contains the noise removal inside it, and running the same cleanup twice is exactly what produces a robotic voice. Picking one switches the other off; nothing is lost either way, because your original recording is untouched. The Export toggle is not a second feature: it is the same enhancement applied in bulk, and it knows what you have already done — it will tell you "already applied to 2 of 3 clips" and only clean the rest. It is not destructive. Turn it off and you are back to your untouched recording — so it is safe to try at each strength and compare. ## Related - [How do I remove background noise?](https://mubert.com/tools/cast/docs/remove-background-noise) - [The AI cleanup made my voice sound robotic — how do I fix it?](https://mubert.com/tools/cast/docs/sounds-robotic) - [Can I change the character of a voice?](https://mubert.com/tools/cast/docs/voice-presets) --- Source: https://mubert.com/tools/cast/docs/speech-enhancer · Cast docs # The AI cleanup made my voice sound robotic — how do I fix it? Lower the strength. Robotic, hollow or underwater artifacts are the signature of noise reduction working too hard — drop from Strong to Medium or Subtle and the voice comes back. Nothing is lost by doing this: your original recording is untouched and the setting is reversible. **Where:** Voice Editor → Polish → strength. Same dial on a clip’s Enhance voice in the Main Editor. ![The strength row of the AI Noise Remover, zoomed: Off, Subtle, Medium, Strong. The fix for a robotic voice is to move left along this row — nothing is lost by doing so, because the original recording is untouched.](https://mubert.com/tools/cast/docs-media/noise-strength.png) This is a real trade-off, not a bug, and it applies to every AI denoiser on the market. Removing noise means deciding which parts of the signal are noise, and the more aggressive that decision, the more of the voice goes with it. A little audible room tone almost always sounds better than a voice with the life processed out of it. If Subtle still sounds processed, turn the cleanup off entirely and judge the raw recording. A quiet, clean-sounding room may not need it at all. ## Related - [How do I remove background noise?](https://mubert.com/tools/cast/docs/remove-background-noise) - [How do I make my voice sound more "studio"?](https://mubert.com/tools/cast/docs/speech-enhancer) - [Can I undo a cleanup — even long after I made it?](https://mubert.com/tools/cast/docs/undo-cleanup) --- Source: https://mubert.com/tools/cast/docs/sounds-robotic · Cast docs # Can I remove echo or room reverb from a recording? Not as a dedicated control — Cast has no de-reverb. Noise removal targets steady noise (hiss, hum, room tone), and it will not pull the echo out of a recording made in a bare, reflective room. **Where:** n/a. ![Steady noise is a separate thing sitting alongside the voice, so it can be told apart and subtracted. An echo is a copy of the voice itself arriving milliseconds later in the same waveform — subtract it and the voice goes too.](https://mubert.com/tools/cast/docs-media/diagrams/remove-echo.svg) This is worth being straight about, because the whole category is vague on it. Echo is your voice arriving twice, so removing it means separating a sound from a copy of itself — that is a much harder problem than removing a steady hiss, and no editor solves it cleanly on an already-recorded file. What genuinely helps is recording into something soft: a room with a rug, curtains and furniture, a mic closer to your mouth, and a body between the mic and the nearest bare wall. Fixing the room beats fixing the file. ## Related - [How do I remove background noise?](https://mubert.com/tools/cast/docs/remove-background-noise) - [How do I make my voice sound more "studio"?](https://mubert.com/tools/cast/docs/speech-enhancer) --- Source: https://mubert.com/tools/cast/docs/remove-echo · Cast docs # Can I change the character of a voice? Yes. Studio Sound presets on the voice lane give it a character in one click — Natural, Radio, Intimate or Interview — and apply to every clip on that lane, including recordings you add later. **Where:** Main Editor → select the voice lane → inspector. ![The Studio Sound card in the voice-lane inspector: a segmented control reading Off, Natural, Radio, Intimate, Interview. The active preset is described underneath, and it applies to every clip on the lane, including new recordings.](https://mubert.com/tools/cast/docs-media/voice-presets.png) Presets are the intended way in. Use them for how the voice should sound; use the Noise Remover and Enhancer for fixing a recording that is genuinely damaged. ## Related - [Can I use EQ and compression?](https://mubert.com/tools/cast/docs/audio-effects) - [How do I make my voice sound more "studio"?](https://mubert.com/tools/cast/docs/speech-enhancer) --- Source: https://mubert.com/tools/cast/docs/voice-presets · Cast docs # Can I use EQ and compression? Yes — an equaliser, compressor, reverb and de-esser can be added per lane, under the Advanced disclosure in the lane inspector. A clip carries only Enhance voice: the rack lives on the lane. **Where:** Main Editor → select a lane → inspector → Advanced. ![The Advanced disclosure opened in the lane inspector, listing the four manual effects that can be added to a lane: equaliser, compressor, reverb and de-esser.](https://mubert.com/tools/cast/docs-media/advanced-fx.png) They sit behind Advanced on purpose: most episodes need "make this voice sound better", which is a preset, not a parametric EQ. Reach for these when you know exactly what you want to change. For a recording with real problems — noise, uneven level — the Noise Remover and Speech Enhancer will get you further than an EQ will. ## Related - [Can I change the character of a voice?](https://mubert.com/tools/cast/docs/voice-presets) - [How do I make my voice sound more "studio"?](https://mubert.com/tools/cast/docs/speech-enhancer) --- Source: https://mubert.com/tools/cast/docs/audio-effects · Cast docs # Music and sound effects Generate music, bring your own, and add SFX. # How do I add my own music? Upload it on the Upload page, tagging it as Music, and it lands in your Library. From there you can put it on a music lane in the Main Editor — either by picking it from the Library, or by using Replace on an existing music clip. **Where:** Upload page → Library → Main Editor. One thing that trips people up: an empty music lane only offers Generate and Library, with no direct upload button. So upload first, then pick your track from the Library. Your own music behaves exactly like generated music once it is on the timeline — it ducks under the voice automatically, and you can set its level and ducking depth. Supported formats are MP3, WAV, M4A, AAC, OGG and FLAC (on Free, WAV and FLAC are held for paid plans). Note that Cast will not license music you brought yourself: the commercial license covers music generated in Cast, not third-party tracks, and clearing those is on you. ## Related - [Why does the music get quieter when someone talks?](https://mubert.com/tools/cast/docs/auto-ducking) - [How do I generate music for my episode?](https://mubert.com/tools/cast/docs/generate-music) - [Is the music licensed for commercial use?](https://mubert.com/tools/cast/docs/licensing) --- Source: https://mubert.com/tools/cast/docs/add-own-music · Cast docs # How do I generate music for my episode? Describe what you want, pick what kind of track it is — a background bed that loops, an intro/outro, or a short sting — and Cast generates an original track for the episode. It is generated for you, not pulled from a catalogue other shows are also using. **Where:** Generate page, or the "+" on a music lane in the Main Editor. **Plan:** Costs credits. The exact price is shown next to the Generate button before you commit. ![The "+" on an empty music lane opens Generate. A style is picked (Voiceover BG) and the length set to thirty seconds; the credit cost is printed beside the Generate button before anything runs — one take reads 5 credits, and asking for a second re-prices it to 10 on the spot. Both takes come back, and adding one drops it onto the timeline already ducking under the voice, marked Auto-ducking −14 dB.](https://mubert.com/tools/cast/docs-media/generate-music.mp4) Because nothing is being searched, "make it calmer" means regenerate rather than go hunting for a different track. If a result is nearly right, re-roll it. Once it is on the timeline it ducks under the voice automatically — you do not draw volume curves. ## Related - [Can I get the music as separate stems?](https://mubert.com/tools/cast/docs/music-stems) - [Can the music be built to fit my episode automatically?](https://mubert.com/tools/cast/docs/match-voice) - [Why does the music get quieter when someone talks?](https://mubert.com/tools/cast/docs/auto-ducking) - [What costs credits?](https://mubert.com/tools/cast/docs/credits-cost) --- Source: https://mubert.com/tools/cast/docs/generate-music · Cast docs # Can the music be built to fit my episode automatically? The length, yes. When you generate a background bed, the Match chip in the Length row sets the track to your episode’s duration, instead of you guessing a length and hoping it fits. **Where:** Generate page → Length → Match. This is the shortest path to a finished episode: clean up the speech first, then generate the bed at Match length — it comes back sized to the result. ## Related - [How do I generate music for my episode?](https://mubert.com/tools/cast/docs/generate-music) - [Why does the music get quieter when someone talks?](https://mubert.com/tools/cast/docs/auto-ducking) --- Source: https://mubert.com/tools/cast/docs/match-voice · Cast docs # Can I get the music as separate stems? Yes. A generated track comes apart into its parts, and you can re-roll one of them — the drums, say — without touching the rest. The parts on offer depend on the track: a gentle voiceover bed may have no drums to swap. **Where:** Swap instruments, on the take you just generated. **Plan:** Downloading stems as files needs Max. ![The Swap instruments panel under a generated take: one chip per part of the track — here Leads and FX — with Leads picked. The price counts only what you picked (one part, 5 credits), and the rest of the track is left alone.](https://mubert.com/tools/cast/docs-media/music-stems.png) Re-rolling one part is the answer to "the track is right but the drums are too busy": keep what works, re-roll only what does not. One re-render covers everything you picked, at the price of a single generation. ## Related - [How do I export stems or individual tracks?](https://mubert.com/tools/cast/docs/export-stems) - [How do I generate music for my episode?](https://mubert.com/tools/cast/docs/generate-music) --- Source: https://mubert.com/tools/cast/docs/music-stems · Cast docs # Is there a library of ready-made music? Yes — a curated library of staff picks you can search in plain words, plus playlists to browse by mood and genre, alongside generation. **Where:** Library → Music. Use the library when you want something immediate and predictable; generate when you want something specific to your show that nobody else is using. ## Related - [How do I generate music for my episode?](https://mubert.com/tools/cast/docs/generate-music) - [Is the music licensed for commercial use?](https://mubert.com/tools/cast/docs/licensing) --- Source: https://mubert.com/tools/cast/docs/curated-library · Cast docs # How do I add sound effects? Describe the sound and Cast generates it — a whoosh, a transition, an ambience. Ask for up to four variants in one run and keep the best. **Where:** SFX page, or the "+" on a lane in the Main Editor. **Plan:** Costs credits, priced per 5 seconds. Shown before you generate. Sound effects are short, so generating a few variants and keeping the best is cheap and normal. ## Related - [What costs credits?](https://mubert.com/tools/cast/docs/credits-cost) - [How do I generate music for my episode?](https://mubert.com/tools/cast/docs/generate-music) --- Source: https://mubert.com/tools/cast/docs/generate-sfx · Cast docs # Is the music licensed for commercial use? Yes — music you generate in Cast comes with a commercial license on Plus and Max, and every export made on those plans carries its own license record. Free music is only free until a platform, a client or a rights holder asks where it came from; here you have the answer on file. (Free and Lite have no commercial license, and music you upload yourself is yours to clear.) **Where:** Exports page → licenses. **Plan:** Commercial license: Plus and Max. ![The Export-ready dialog listing the download contents: the full mix as MP3, two caption files, and two Creator licenses as PDFs — one per generated music track, bundled into the same ZIP.](https://mubert.com/tools/cast/docs-media/export-licenses.png) Publish on podcast platforms, YouTube or a client channel without waiting to find out whether the track was clear. Each export made on a licensed plan comes with its license record — that is the document you point at if anyone asks. The one thing to decide up front: if your show carries ads or sponsorship, generate the music on Plus or above, and the license travels with the export. ## Related - [What do the plans include?](https://mubert.com/tools/cast/docs/plans-and-limits) - [How do I add my own music?](https://mubert.com/tools/cast/docs/add-own-music) - [How do I generate music for my episode?](https://mubert.com/tools/cast/docs/generate-music) - [How do I export my episode as an MP3 or WAV?](https://mubert.com/tools/cast/docs/export-audio) --- Source: https://mubert.com/tools/cast/docs/licensing · Cast docs # Mixing The Main Editor: lanes, clips, and music that gets out of the way of the voice. # Why does the music get quieter when someone talks? That is auto-ducking. The music drops under the voice and comes back in the gaps, automatically — and because the duck follows your actual voice clips, it stays correct after you cut. Trim a minute of filler and the music re-balances itself. You never draw a volume envelope by hand. You can set how far it drops, or switch it off. **Where:** Main Editor → select the music lane → inspector. ![The music level drops while the voice is speaking and rises back in the gap between sentences. The curve follows your voice clips, so it stays correct after you cut — you never draw it by hand. In the Main Editor the control sits on the music lane, shown as an Auto-ducking chip with its depth in dB.](https://mubert.com/tools/cast/docs-media/diagrams/what-is-ducking.svg) That is the part other tools leave to you: in an editor where the music is a track you imported, every edit to the speech means going back and redrawing the automation around it. If the music still feels loud under the voice, deepen the ducking and lower the music level — not the master, which just makes the whole episode quieter. ## Related - [The music is too loud under my voice — what do I fix?](https://mubert.com/tools/cast/docs/music-too-loud) - [How do I generate music for my episode?](https://mubert.com/tools/cast/docs/generate-music) - [What LUFS should my podcast be?](https://mubert.com/tools/cast/docs/loudness-lufs) --- Source: https://mubert.com/tools/cast/docs/auto-ducking · Cast docs # How does the timeline work? The Main Editor is lane-based: voice on one lane, music and sound effects on others. Clips can be dragged, split, trimmed and replaced, and your chapters run along the top. **Where:** Main Editor. ![The Main Editor: a lane-based timeline with the voice on one lane, music and sound effects on others, chapters along the top and the playhead running through them.](https://mubert.com/tools/cast/docs-media/main-editor.png) Edits from the Voice Editor are already applied here — the voice clip you see is the cleaned-up one, not the raw recording. To swap a clip's audio without rebuilding the arrangement, use Replace on the lane head: you can replace it from the Library, from a file you upload, or by regenerating it. ## Related - [Which editor do I use — Voice or Main?](https://mubert.com/tools/cast/docs/which-editor) - [Why does the music get quieter when someone talks?](https://mubert.com/tools/cast/docs/auto-ducking) - [How do I add my own music?](https://mubert.com/tools/cast/docs/add-own-music) --- Source: https://mubert.com/tools/cast/docs/timeline-clips · Cast docs # Exporting and publishing Platform-matched loudness, stems, captions and chapters — and the license record that goes with the episode. # How do I export my episode as an MP3 or WAV? Open Export, pick the platform you are publishing to, and choose MP3 or WAV. Cast renders the mix and normalizes it to that platform's loudness. Exporting costs no credits — on any plan, however many times you do it. What changes with your plan is the quality you can reach, not what you can do. **Where:** Export page. **Plan:** Free: MP3 128k. Lite: MP3 320k. Plus: WAV. Max: broadcast-grade. ![The Export dialog: choose a platform preset, then MP3 or WAV. Toggles for stems, voice enhancement and the Creator licence PDF, with the estimated file size and render time shown before you commit.](https://mubert.com/tools/cast/docs-media/export-modal.png) MP3 is compressed and is what you publish: it is a fraction of the size and every podcast platform re-encodes it anyway. WAV is uncompressed and is what you hand to someone else who will do more work on the audio — a mastering engineer, a video editor, an archive. If the episode uses music you generated on a licensed plan, the export carries its license record with it — see Is the music licensed for commercial use? ## Related - [What LUFS should my podcast be?](https://mubert.com/tools/cast/docs/loudness-lufs) - [Why can't I download a WAV file?](https://mubert.com/tools/cast/docs/cant-export-wav) - [How do I export stems or individual tracks?](https://mubert.com/tools/cast/docs/export-stems) - [Is the music licensed for commercial use?](https://mubert.com/tools/cast/docs/licensing) --- Source: https://mubert.com/tools/cast/docs/export-audio · Cast docs # What LUFS should my podcast be? Apple Podcasts wants −16 LUFS; Spotify and YouTube want −14 LUFS; broadcast (EBU R128) wants −23 LUFS. Most tools make you type that number in. Cast ships the target with the platform — pick where you are publishing and it normalizes to that, with true peak at −1 dB. (Master is the exception: no normalization, no ceiling — untouched on purpose.) **Where:** Export page → preset. ![The Export dialog with all six loudness presets side by side: Spotify Podcast MP3 320k at -14 LUFS, Apple Podcasts -16 LUFS, Spotify/Apple Music -14, YouTube -14, Broadcast EBU R128 WAV 48k at -23, and Master with no normalization. Below them: format, stems, Creator licence, and the estimated duration, size and render time.](https://mubert.com/tools/cast/docs-media/export-modal.png) Loudness normalization is why one podcast sounds as loud as the next one in a listener's feed. Every platform re-levels what you upload; hitting its target yourself means it does not have to, so your episode arrives sounding the way you mixed it instead of being turned down by an algorithm. If someone else is going to master the audio, export with Master, which applies no normalization at all. The MP3 presets render at 320k on Lite and above; on Free the same presets render at 128k — the loudness target is identical. - **Spotify Podcast** — −14 LUFS · MP3 320k - **Apple Podcasts** — −16 LUFS · MP3 320k - **Spotify / Apple Music** — −14 LUFS · MP3 320k - **YouTube** — −14 LUFS · MP3 320k - **Broadcast (EBU R128)** — −23 LUFS · WAV 48 kHz · Max plan - **Master** — No normalization · WAV 48 kHz, 16-bit ## Related - [How do I export my episode as an MP3 or WAV?](https://mubert.com/tools/cast/docs/export-audio) - [The music is too loud under my voice — what do I fix?](https://mubert.com/tools/cast/docs/music-too-loud) --- Source: https://mubert.com/tools/cast/docs/loudness-lufs · Cast docs # Why can't I download a WAV file? There is no watermark on any plan — your episode is your episode, even on Free. What a plan sets is the quality ceiling: Free and Lite export MP3 (128k and 320k), WAV and lossless unlock on Plus, and broadcast-grade output (EBU R128) is on Max. **Where:** Export page. Pricing page to change plan. **Plan:** WAV: Plus or Max. ![The Export dialog showing the format choice: MP3, lossy and small, next to WAV, lossless and large. Free and Lite export MP3; WAV unlocks on Plus. There is no watermark option anywhere in the dialog.](https://mubert.com/tools/cast/docs-media/export-modal.png) Nothing is held hostage. Every plan exports a finished, publishable file as many times as you like, at no credit cost. ## Related - [What do the plans include?](https://mubert.com/tools/cast/docs/plans-and-limits) - [How do I export my episode as an MP3 or WAV?](https://mubert.com/tools/cast/docs/export-audio) --- Source: https://mubert.com/tools/cast/docs/cant-export-wav · Cast docs # How do I export stems or individual tracks? Turn on stems in the export dialog and you get a ZIP of the separate parts of the mix alongside the full episode. Available on Max. **Where:** Export page → stems. **Plan:** Max. ![The Export dialog with the stems toggle: each track as its own file, with its effects, plus the full mix, all in one ZIP.](https://mubert.com/tools/cast/docs-media/export-modal.png) The music arrives as stems, not a flattened bake — so an engineer or a DAW picks up exactly where you left off, and you are never locked in. ## Related - [Can I get the music as separate stems?](https://mubert.com/tools/cast/docs/music-stems) - [Why can't I download a WAV file?](https://mubert.com/tools/cast/docs/cant-export-wav) - [Is the music licensed for commercial use?](https://mubert.com/tools/cast/docs/licensing) --- Source: https://mubert.com/tools/cast/docs/export-stems · Cast docs # Where do my past exports live? On the Exports page. Every export you have made is listed there and can be downloaded again — with its license PDFs bundled in, if it was made on a licensed plan — so losing the downloaded file never means re-rendering the episode. **Where:** Exports — in the main navigation. Captions, chapters and the transcript can be downloaded again from there too, in the same formats the export dialog offers. This is also where the license records live: each export made on Plus or Max lists its certificates next to the file they cover. ## Related - [How do I export my episode as an MP3 or WAV?](https://mubert.com/tools/cast/docs/export-audio) - [Is the music licensed for commercial use?](https://mubert.com/tools/cast/docs/licensing) - [How do I export captions or subtitles?](https://mubert.com/tools/cast/docs/captions-subtitles) --- Source: https://mubert.com/tools/cast/docs/where-are-my-exports · Cast docs # How do I export captions or subtitles? Captions come from your edited transcript and download as SRT, VTT or plain text — so the words you cut do not come back in the subtitles. **Where:** Export page, and the transcript toolbar in the Voice Editor. That the captions match the edited audio sounds obvious, and it is exactly what breaks when transcription and editing live in different tools: you cut a sentence, and it is still sitting in the SRT you uploaded to YouTube. Cast is audio-first by design — there is no video timeline to fight — so captions ship as sidecar files that match the edited audio exactly, and you upload them alongside the episode. ## Related - [How do I export chapters with timestamps?](https://mubert.com/tools/cast/docs/export-chapters) - [How do I fix a word the transcription got wrong?](https://mubert.com/tools/cast/docs/correct-transcript) --- Source: https://mubert.com/tools/cast/docs/captions-subtitles · Cast docs # How do I export chapters with timestamps? Chapters export as a timestamp list in two formats: YouTube description chapters, and Podcasting 2.0 chapters for podcast apps. The full transcript can also be exported as JSON. **Where:** Export page, and the transcript toolbar. Paste the YouTube list straight into a video description and it becomes clickable chapters. The Podcasting 2.0 file goes to podcast hosts that support chapter markers. ## Related - [How do I add chapters with timestamps?](https://mubert.com/tools/cast/docs/chapters) - [How do I export captions or subtitles?](https://mubert.com/tools/cast/docs/captions-subtitles) --- Source: https://mubert.com/tools/cast/docs/export-chapters · Cast docs # Plans and credits What each plan includes, and what actually costs credits. # What do the plans include? Editing is never metered and never watermarked: transcription, filler and pause removal, noise removal, captions and audio export cost zero credits on every plan, including Free. Credits buy generation (music and SFX); the uploads allowance caps how much audio you bring in. Free: 100 credits, 60 min, MP3 128k. Lite: 700, 10 hrs, MP3 320k. Plus: 2,750, 25 hrs, WAV, commercial license. Max: 6,500, unlimited, broadcast (EBU R128), stems. **Where:** Pricing page. ![Only generation — music and sound effects — costs credits. Transcription, cutting filler words, trimming pauses, noise removal, captions and audio export are free on every plan including Free, with no watermark.](https://mubert.com/tools/cast/docs-media/diagrams/plans-metering.svg) Two things do the gating, and neither of them is the editing: how much audio you can bring in per month, and how many credits you have for generating music and sound effects. So a Free account is a whole workflow, not a teaser — you can transcribe, clean and export a finished episode without spending a credit. - **Free** — 100 credits/mo · 60 min uploads/mo · MP3 128k · no commercial license - **Lite** — 700 credits/mo · 10 hrs uploads/mo · MP3 320k - **Plus** — 2,750 credits/mo · 25 hrs uploads/mo · WAV · commercial license - **Max** — 6,500 credits/mo · unlimited uploads · broadcast (EBU R128) · stems · commercial license ## Related - [What costs credits?](https://mubert.com/tools/cast/docs/credits-cost) - [Is the music licensed for commercial use?](https://mubert.com/tools/cast/docs/licensing) - [Why can't I download a WAV file?](https://mubert.com/tools/cast/docs/cant-export-wav) --- Source: https://mubert.com/tools/cast/docs/plans-and-limits · Cast docs # What costs credits? Only generation — music and sound effects. Transcription, every kind of transcript editing, noise removal, captions and exporting are free on every plan. The exact price is shown next to the Generate button before you spend anything. **Where:** Balance is in the top bar; the cost sits beside every generate button. Music is priced by length: a base cost covers the first three minutes, and each additional minute adds to it. Sound effects are priced per five seconds started. Credits come with your plan each month. More credits come from moving up a plan — there is no separate credit pack to buy. - **Music** — Base cost covers the first 3 minutes, then per extra minute. - **SFX** — Per started 5 seconds. - **Free** — Transcription · transcript editing · noise removal · captions · audio export. ## Related - [What happens when I run out of credits?](https://mubert.com/tools/cast/docs/out-of-credits) - [What do the plans include?](https://mubert.com/tools/cast/docs/plans-and-limits) - [How do I generate music for my episode?](https://mubert.com/tools/cast/docs/generate-music) --- Source: https://mubert.com/tools/cast/docs/credits-cost · Cast docs # What happens when I run out of credits? Generation stops; everything else keeps working. You can still edit, transcribe, clean up audio, add captions and export. Credits refresh monthly, or you can upgrade for a bigger allowance. **Where:** Pricing page. Nothing you already generated is taken away. The tracks stay in your projects and in your library. ## Related - [What costs credits?](https://mubert.com/tools/cast/docs/credits-cost) - [What do the plans include?](https://mubert.com/tools/cast/docs/plans-and-limits) --- Source: https://mubert.com/tools/cast/docs/out-of-credits · Cast docs # Account and limits Subscriptions, limits, and where Cast deliberately stops — and what picks up from there. # How do I sign in — and what if I forgot my password? Sign in with Google, or with your email and a password. Forgot the password? The Forgot password link on the sign-in page emails you a reset link — follow it and set a new one. **Where:** The sign-in page; Forgot password lives under the password field. The two methods reach the same account only if they use the same email address. If projects seem to be missing after signing in, the first thing to check is which email you signed in with. The reset link lands in your inbox within a minute or two. If it does not, check spam — and make sure you typed the address you registered with, because for safety the page will not tell you whether an address exists. ## Related - [How do I cancel or change my plan?](https://mubert.com/tools/cast/docs/cancel-subscription) - [How do I delete my account?](https://mubert.com/tools/cast/docs/delete-account) --- Source: https://mubert.com/tools/cast/docs/sign-in-password · Cast docs # How do I delete my account? In Settings. Deleting the account is permanent: it removes the account and every project in it, and there is no undo and no recovery window — so export anything you want to keep first. **Where:** Settings → Delete account. Cast asks you to confirm before it does anything, and the confirmation says exactly what will happen: the account and all projects go, irreversibly. If the goal is to stop paying rather than to erase everything, cancel the subscription instead — you drop to the Free plan and your projects stay. ## Related - [How do I cancel or change my plan?](https://mubert.com/tools/cast/docs/cancel-subscription) - [I hit a limit — what now?](https://mubert.com/tools/cast/docs/limit-reached) --- Source: https://mubert.com/tools/cast/docs/delete-account · Cast docs # How do I cancel or change my plan? Both happen in the billing portal, which you reach from Settings. Cancelling takes effect at the end of the period you have already paid for — you keep your plan until then. **Where:** Settings → billing. Upgrading and downgrading go through the same portal, and that is also where you update your card and find your invoices. ## Related - [What do the plans include?](https://mubert.com/tools/cast/docs/plans-and-limits) - [What happens when I run out of credits?](https://mubert.com/tools/cast/docs/out-of-credits) --- Source: https://mubert.com/tools/cast/docs/cancel-subscription · Cast docs # Can I pause a recording in progress? Record in takes. A take runs start-to-stop — there is no pause and resume — and each one lands as its own clip you can reorder, trim and restore independently on the timeline. **Where:** Record page. You do get a device picker and a live input level meter before and during the take, so you can check you are recording from the right microphone and are not clipping. One thing to know before you start: a take lives in the browser tab until you stop and save it. If you close the tab mid-recording, that take is gone — there is nothing to recover. On a long recording, stopping and saving in sections is the safer habit. ## Related - [How do I get my audio into Cast?](https://mubert.com/tools/cast/docs/import-audio) - [How does the timeline work?](https://mubert.com/tools/cast/docs/timeline-clips) --- Source: https://mubert.com/tools/cast/docs/can-i-pause-recording · Cast docs # Can Cast publish my podcast to Spotify or Apple? Cast finishes the episode; your host distributes it. Cast is deliberately host-agnostic — no RSS, no lock-in — and the file it hands you is already normalized to the target platform's loudness, so it drops into any podcast host without a re-master. You leave with the host-ready bundle: the exported MP3 at the right loudness, your chapters file, and your transcript. That is why picking the right loudness preset matters here. It is the last chance to get the level right before the file leaves. ## Related - [What LUFS should my podcast be?](https://mubert.com/tools/cast/docs/loudness-lufs) - [How do I export chapters with timestamps?](https://mubert.com/tools/cast/docs/export-chapters) - [How do I export my episode as an MP3 or WAV?](https://mubert.com/tools/cast/docs/export-audio) --- Source: https://mubert.com/tools/cast/docs/publish-podcast · Cast docs # I hit a limit — what now? Nothing you already made is ever taken away, hidden or deleted because you crossed a limit — not even if you move down to a smaller plan. Limits apply only when you create something new: at your project limit (Free holds five projects), delete a finished project or move up a plan; at your monthly upload allowance, it resets with your next billing month. **Where:** Pricing page. Nothing you already made is taken away, hidden or deleted because you crossed a limit — including if you move down to a smaller plan. Your projects and audio stay where they are; you simply cannot add new ones until you are back under the line. Storage works the same way: past the ceiling, new uploads are refused, and the work you already have is untouched. ## Related - [What do the plans include?](https://mubert.com/tools/cast/docs/plans-and-limits) - [What happens when I run out of credits?](https://mubert.com/tools/cast/docs/out-of-credits) - [How do I cancel or change my plan?](https://mubert.com/tools/cast/docs/cancel-subscription) --- Source: https://mubert.com/tools/cast/docs/limit-reached · Cast docs # Where do I find the keyboard shortcuts? Press ? in either editor and the shortcut sheet opens, showing the keys for the editor you are actually in — they differ between the Voice Editor and the Main Editor. **Where:** Voice Editor or Main Editor — press ?. ![The keyboard cheat sheet in the Voice Editor, opened with the ? key: playback, transcript actions such as assigning a speaker with Shift-S and making a chapter with Cmd-Shift-M, editing, and workflow keys.](https://mubert.com/tools/cast/docs-media/shortcuts-voice.png) The ones worth learning first are the ones you will use on every episode: Space to play and pause, ⌘Z to undo, and in the transcript, Enter to split a paragraph and ⇧S to assign a speaker. ## Related - [How do I edit the transcript — what can I actually do?](https://mubert.com/tools/cast/docs/edit-transcript) - [Which editor do I use — Voice or Main?](https://mubert.com/tools/cast/docs/which-editor) --- Source: https://mubert.com/tools/cast/docs/keyboard-shortcuts · Cast docs # Can I bring my own transcript? No. Cast transcribes the audio itself, and the transcript has to be its own — every word is tied to a moment in the recording, which is what lets you edit the audio by editing the text. ![Every word Cast transcribes is pinned to a moment in the audio, and that anchoring is what lets deleting text delete sound. An imported transcript is text sitting next to the audio with nothing pointing into it, so deleting a sentence would cut nothing.](https://mubert.com/tools/cast/docs-media/diagrams/import-transcript.svg) An imported transcript would just be text: without word-level timing bound to your audio, deleting a sentence could not delete the sound of it. If the transcription got words wrong, fix them in place with Correct or Find & Replace. Re-transcribing is free, so pinning the right language and running it again is often the faster fix. ## Related - [How do I fix a word the transcription got wrong?](https://mubert.com/tools/cast/docs/correct-transcript) - [Which languages can Cast transcribe?](https://mubert.com/tools/cast/docs/transcription-languages) --- Source: https://mubert.com/tools/cast/docs/import-transcript · Cast docs # Troubleshooting When something does not work, or does not sound right. # The music is too loud under my voice — what do I fix? Lower the music lane's level and deepen the ducking. Do not reach for the master: ducking controls how far the music drops while someone is speaking, while the master sets the level of the whole episode, so lowering it just makes everything quieter together. **Where:** Main Editor → the music lane. ![Ducking is the amount the music drops while someone is speaking. Deepening it lowers the music under the voice without touching the master, which would only make the whole episode quieter.](https://mubert.com/tools/cast/docs-media/diagrams/what-is-ducking.svg) If it still sounds hot after export, check which preset you exported with. Master applies no loudness normalization, so a mix that was already loud stays loud. Pick the platform preset instead. ## Related - [Why does the music get quieter when someone talks?](https://mubert.com/tools/cast/docs/auto-ducking) - [What LUFS should my podcast be?](https://mubert.com/tools/cast/docs/loudness-lufs) --- Source: https://mubert.com/tools/cast/docs/music-too-loud · Cast docs # Audio in an old project will not play Reload the page. Audio links expire after a while for security, and reloading fetches fresh ones — nothing has been lost. This shows up on projects you come back to after a long gap. It is a link problem, not a data problem: your recordings, your edits and your generated tracks are all still there. ## Related - [How do I get my audio into Cast?](https://mubert.com/tools/cast/docs/import-audio) --- Source: https://mubert.com/tools/cast/docs/audio-wont-play · Cast docs # My audio was not transcribed Transcription starts on its own once an upload completes, so no transcript usually means the upload did not finish. Check the file is a supported audio format — MP3, WAV, M4A, AAC, OGG or FLAC — and upload it again. Long files take longer than you might expect. Give it a minute before assuming it has failed. If the language was detected wrong, you do not need to re-upload — pin the language on the file and transcribe again. It costs nothing. ## Related - [How do I get my audio into Cast?](https://mubert.com/tools/cast/docs/import-audio) - [Which languages can Cast transcribe?](https://mubert.com/tools/cast/docs/transcription-languages) --- Source: https://mubert.com/tools/cast/docs/no-transcript · Cast docs # My file will not upload Four usual causes, each with its own message: the file is over your plan’s per-file limit (200 MB and 60 minutes on Free, 1 GB and 2 hours on paid), it is a WAV or FLAC on the Free plan (lossless uploads are paid), the format is not a supported audio format (video is rejected), or you have used up your monthly upload allowance, which resets with your billing month. For an oversized file — and for a lossless file on Free — the same move fixes both: re-export it as a 320k MP3. An hour of audio lands well under every limit, and transcription and editing work the same on it. For a video, extract the audio track and upload that. And a file rejected as empty really is empty — zero bytes of audio — so re-export it from wherever it came from. ## Related - [How do I get my audio into Cast?](https://mubert.com/tools/cast/docs/import-audio) - [What do the plans include?](https://mubert.com/tools/cast/docs/plans-and-limits) - [I hit a limit — what now?](https://mubert.com/tools/cast/docs/limit-reached) --- Source: https://mubert.com/tools/cast/docs/file-wont-upload · Cast docs # The recorder cannot see my microphone That is the browser withholding permission, not Cast. Click the padlock or settings icon next to the address bar, allow the microphone for this site, and reload — the device picker on the Record page then lists everything the browser can see. If the right mic is listed but silent, check the input meter while you speak: no movement means the operating system is pointing at a different device, or the mic is muted at the hardware level. One thing to know: the device cannot be switched mid-take — pick the mic before you press record. ## Related - [Can I pause a recording in progress?](https://mubert.com/tools/cast/docs/can-i-pause-recording) - [How do I get my audio into Cast?](https://mubert.com/tools/cast/docs/import-audio) --- Source: https://mubert.com/tools/cast/docs/mic-not-detected · Cast docs # My export failed or is stuck Start it again — exports are re-runnable and a failed one costs you nothing. Long episodes genuinely take a while to render, so a slow export is not necessarily a stuck one. If it fails repeatedly, try a different format or a shorter section to work out whether the problem is the length or something in the mix. ## Related - [How do I export my episode as an MP3 or WAV?](https://mubert.com/tools/cast/docs/export-audio) --- Source: https://mubert.com/tools/cast/docs/export-stuck · Cast docs # I broke my project — how do I get an earlier version back? Open the clock icon in the top bar → "Version history…", pick the checkpoint from before things went wrong, and press Restore. Restoring never deletes anything — the state you are leaving is checkpointed too, so you can always come back forward. **Where:** Either editor → clock icon in the top bar → Version history… Look for the semantic labels first: Cast takes a checkpoint right before every export, Enhance voice, re-transcribe and bulk cleanup, so "Before re-transcribe" is usually exactly the version you want. Between those, an "Editing checkpoint" is cut roughly every half hour of active work. Not sure which checkpoint is the right one? Use Duplicate instead of Restore — it opens the checkpoint as a new project so you can look around without touching the current state of this one. For a single bad edit, plain undo (⌘Z) inside the session is still the fastest way back; version history is for the cases undo cannot reach — after a reload, or when a bulk operation changed too much at once. ## Related - [How does version history work?](https://mubert.com/tools/cast/docs/version-history) - [Is my work saved automatically?](https://mubert.com/tools/cast/docs/does-cast-save) --- Source: https://mubert.com/tools/cast/docs/restore-version · Cast docs # How Cast compares Where Cast fits against the other tools — including when to use one of them instead. # Cast vs Descript Both let you edit audio by editing a transcript. Descript is a broad all-in-one that also does video and screen recording, and it has automatic speaker detection. Cast is a finisher, not an all-in-one: audio only, no video timeline, no stock-music rabbit hole. Descript gives you a library to search and a fader to ride; Cast generates the track for the episode, hands you the stems, ducks it under the voice on its own, and clears it for commercial use. Pick Descript if the deliverable is video. On speakers, Descript will guess who is talking; Cast asks you to say so once (⇧S) and then remembers. On a two-mic recording the guess is usually right; on one shared mic no tool is, and a wrong guess costs more to unpick than a right label costs to type. Pick Cast if the output is audio and the music is part of the job. In Descript, scoring an episode means finding a track and riding its volume around the speech. In Cast the music is generated for the episode, in stems, and it ducks under the voice on its own — and on Plus and Max it comes with a commercial license, so a sponsored show does not become a rights question later. The difference that shows up on a bad day: Cast never writes over your recording. Every cut, mute, filler removal and AI enhancement is a layer on top of an untouched original, and one Restore button puts it back — including after export. There is no "save a copy first" ritual, because there is nothing to lose. - **Both** — Edit audio by editing the transcript. Filler-word and pause removal. AI voice cleanup. - **Descript only** — Video and screen recording. Automatic speaker detection. Overdub-style voice tools. - **Cast only** — Generative music per episode, in stems, with a commercial license. Automatic ducking. Platform loudness presets. Non-destructive — nothing is ever overwritten. ## Related - [Cast vs Riverside](https://mubert.com/tools/cast/docs/mucast-vs-riverside) - [How do I separate two speakers in one audio track?](https://mubert.com/tools/cast/docs/separate-speakers) - [Why does the music get quieter when someone talks?](https://mubert.com/tools/cast/docs/auto-ducking) --- Source: https://mubert.com/tools/cast/docs/mucast-vs-descript · Cast docs # Cast vs Riverside Riverside is a remote recording studio first — it records each guest locally at high quality and then gives you an editor. Cast starts where recording ends. They are not substitutes: Riverside sells you a studio, Cast hands you a finished episode. If you record interviews over the internet, use both — Riverside to capture, Cast to finish. On speakers, the two are closer than the marketing suggests. Riverside labels one speaker per recorded track, and its own docs say plainly that it cannot tell two people apart on one microphone — the advice is to give each person their own mic. Cast asks you to label speakers by hand, and gives the same advice. Where Cast pulls ahead is everything after the cut: music generated for the episode that ducks itself and carries a commercial license, loudness presets that actually target Spotify and Apple, stems export, and an original recording that is never overwritten — so any cleanup you regret is one Restore away. - **Riverside only** — Remote multi-guest recording, captured locally per participant. Video. - **Cast only** — Generative music with automatic ducking, in stems, commercially licensed. Loudness presets. Non-destructive — nothing is ever overwritten. - **Neither** — Reliably separating two people recorded on one microphone. Use one mic per person. ## Related - [Cast vs Descript](https://mubert.com/tools/cast/docs/mucast-vs-descript) - [How do I separate two speakers in one audio track?](https://mubert.com/tools/cast/docs/separate-speakers) - [How do I get my audio into Cast?](https://mubert.com/tools/cast/docs/import-audio) --- Source: https://mubert.com/tools/cast/docs/mucast-vs-riverside · Cast docs # Cast vs Adobe Podcast Enhance Adobe Podcast Enhance is a single-purpose cleanup tool: you give it a file, it gives you a cleaner-sounding file. Cast does the same cleanup with two things a one-shot tool cannot give you — a strength dial, and an original that was never overwritten. The complaint people have with one-shot enhancers is that the result can come back robotic or hollow, and there is no dial to back it off — you take what you are given. That hollowness is the trade a one-shot pass makes to remove a room. Cast's noise removal and speech enhancer both run at Subtle, Medium or Strong, and switching them off restores your untouched recording, so you can hear the trade and decide how much of it you want. Beyond cleanup, Enhance does not edit, does not do music, and does not export to a loudness target. If all you need is a quick pass over one file, it is a reasonable free tool. If you are making episodes, it is one feature, not a workflow. - **Both** — AI cleanup of a noisy voice recording. - **Adobe only** — A free one-shot pass, with nothing to learn. - **Cast only** — A strength dial instead of take-what-you-get. Transcript editing. Generative music with a commercial license. Loudness presets. Non-destructive — the original is always there. ## Related - [The AI cleanup made my voice sound robotic — how do I fix it?](https://mubert.com/tools/cast/docs/sounds-robotic) - [How do I remove background noise?](https://mubert.com/tools/cast/docs/remove-background-noise) - [How do I make my voice sound more "studio"?](https://mubert.com/tools/cast/docs/speech-enhancer) --- Source: https://mubert.com/tools/cast/docs/mucast-vs-adobe-podcast · Cast docs # Cast vs a stock music library A stock library (Epidemic Sound, Artlist, Uppbeat and the rest) sells you a subscription to a catalogue: you search it, audition tracks, find one that nearly fits, and trim it to length. Cast generates the track for this episode instead — sized to your voice, delivered in stems, ducking itself under the speech, and on Plus and Max carrying a commercial license. The catalogue is the whole problem. A track that "nearly fits" is one you then have to cut, loop or fade to length, and ride the volume on around every sentence — and the fit never gets better than nearly, because the track was not made for your episode. Then there is the part nobody enjoys: the rights. Stock libraries license by subscription, which means the clearance on an episode you published last year can depend on a subscription you are still paying for. And the free tiers of these libraries are where "free music" gets people into trouble — it is free until a platform, a client or a rights holder asks where it came from. Cast generates the music inside the same place you edit the episode, so the track knows about your edits: cut a minute of speech and it re-balances. Each export made on a licensed plan comes with its license record — a document you can point at. When a library is still the right answer: you want a specific, recognizable, human-composed piece, or a named artist. Generation does not give you that, and Cast ships a curated library for exactly this reason. - **Both** — A large amount of music you are allowed to use in a monetized show. - **Stock library only** — Human-composed catalogue tracks, named artists, a specific song you already have in mind. - **Cast only** — Music generated for this episode, sized to your voice. Stems, not a flat file. Automatic ducking that survives your edits. A license record attached to the export. ## Related - [Is the music licensed for commercial use?](https://mubert.com/tools/cast/docs/licensing) - [How do I generate music for my episode?](https://mubert.com/tools/cast/docs/generate-music) - [Is there a library of ready-made music?](https://mubert.com/tools/cast/docs/curated-library) - [Can I get the music as separate stems?](https://mubert.com/tools/cast/docs/music-stems) --- Source: https://mubert.com/tools/cast/docs/cast-vs-stock-music · Cast docs # Glossary The audio words that get thrown at podcasters, in plain language. # What is LUFS? LUFS (Loudness Units Full Scale) measures how loud something actually sounds to a person, averaged over time — unlike peak level, which only measures the single loudest instant. Podcast platforms use it as their delivery target: Apple Podcasts wants −16 LUFS, Spotify and YouTube −14 LUFS. ![Three episodes in one feed at different loudness. The quiet one makes the listener reach for the volume; the loud one gets turned down by the platform. Hitting the target means yours arrives sounding the way you mixed it.](https://mubert.com/tools/cast/docs-media/diagrams/what-is-lufs.svg) The reason it exists: a quiet recording with one loud door slam has a high peak but is still a quiet recording. LUFS matches how loudness is perceived, so it is the number platforms can meaningfully normalize against. A lower number is quieter — and because these are negative, −23 LUFS is quieter than −14 LUFS. In Cast you never type this number. Pick your platform on export and the target is applied. ## Related - [What LUFS should my podcast be?](https://mubert.com/tools/cast/docs/loudness-lufs) - [What is true peak, and why −1 dB?](https://mubert.com/tools/cast/docs/what-is-true-peak) - [What does normalizing audio mean?](https://mubert.com/tools/cast/docs/what-is-normalization) --- Source: https://mubert.com/tools/cast/docs/what-is-lufs · Cast docs # What is true peak, and why −1 dB? True peak is the highest level your audio will actually reach once it is converted back to sound — including peaks that appear between the digital samples. Leaving 1 dB of headroom (a −1 dBTP ceiling) stops those hidden peaks from distorting when a platform re-encodes your file to MP3 or AAC. ![A waveform that touches the ceiling gets flattened tops and distorts once a platform re-encodes it, because peaks appear between the digital samples. Leaving 1 dB of headroom gives those peaks somewhere to go.](https://mubert.com/tools/cast/docs-media/diagrams/what-is-true-peak.svg) This is why a file that looked fine in your editor can come back crunchy from a streaming platform: the encoder introduced peaks that were not in your version. The headroom is insurance against that. Every platform preset in Cast targets −1 dBTP. The one exception is Master, which applies no processing at all — true peak included — because someone else is going to master it. ## Related - [What is LUFS?](https://mubert.com/tools/cast/docs/what-is-lufs) - [What LUFS should my podcast be?](https://mubert.com/tools/cast/docs/loudness-lufs) --- Source: https://mubert.com/tools/cast/docs/what-is-true-peak · Cast docs # What does normalizing audio mean? Normalizing means adjusting the whole recording up or down so it lands on a target level. It changes loudness, not balance — it will not fix a quiet guest against a loud host, because it moves everything by the same amount. ![Normalizing lifts the whole recording to a target level, so a quiet guest and a loud host both move up by the same amount. The gap between them is unchanged — normalizing fixes loudness, not balance.](https://mubert.com/tools/cast/docs-media/diagrams/what-is-normalization.svg) That distinction catches people out constantly. If two people are at different levels, normalizing the mix does nothing to the gap between them; you have to fix the levels of the two voices relative to each other first. ## Related - [What is LUFS?](https://mubert.com/tools/cast/docs/what-is-lufs) - [What LUFS should my podcast be?](https://mubert.com/tools/cast/docs/loudness-lufs) --- Source: https://mubert.com/tools/cast/docs/what-is-normalization · Cast docs # What is ducking (or sidechain)? Ducking is automatically lowering the music whenever someone speaks, and letting it come back up when they stop. It is what makes a music bed sit under a voice instead of fighting it — and doing it by hand means drawing a volume curve around every sentence. ![The music level drops while the voice is speaking and rises back in the gap between sentences. The curve follows the voice clips, so it stays correct after you cut — you never draw it by hand.](https://mubert.com/tools/cast/docs-media/diagrams/what-is-ducking.svg) You have heard it in every radio ad and podcast intro. When it is done well you do not notice it; when it is missing, the voice sounds buried. Cast ducks automatically, following your actual voice clips, so it stays correct even after you cut the speech. ## Related - [Why does the music get quieter when someone talks?](https://mubert.com/tools/cast/docs/auto-ducking) - [The music is too loud under my voice — what do I fix?](https://mubert.com/tools/cast/docs/music-too-loud) --- Source: https://mubert.com/tools/cast/docs/what-is-ducking · Cast docs # What are stems? Stems are the separate parts of a finished mix, kept as individual files — the voice on one, the drums on another, the bass on another. A stereo mix is those parts already blended together and no longer separable. ![Stems are the parts of a mix kept as separate files — voice, music, SFX — so any one of them can be changed alone. A stereo mix is those parts already blended and no longer separable.](https://mubert.com/tools/cast/docs-media/diagrams/what-are-stems.svg) They matter because they are what lets someone else keep working. Handing an engineer a finished mix means they can only adjust the whole thing; handing them stems means they can turn the music down without touching the voice. ## Related - [How do I export stems or individual tracks?](https://mubert.com/tools/cast/docs/export-stems) - [Can I get the music as separate stems?](https://mubert.com/tools/cast/docs/music-stems) --- Source: https://mubert.com/tools/cast/docs/what-are-stems · Cast docs # What does non-destructive editing mean? It means your edits are stored as instructions layered over the original recording, which is never overwritten. The practical consequence: any cut can be undone at any time, including one you made weeks ago, and including after you have exported. ![In Cast the original recording is never overwritten: edits are stored as removable layers on top of it, so Restore can lift any one of them off, even after export. Destructive editing bakes each change into the file, and an over-aggressive pass is unrecoverable.](https://mubert.com/tools/cast/docs-media/diagrams/what-is-non-destructive.svg) The opposite — destructive editing — bakes each change into the audio file. That is how you end up with a recording you cannot recover after an over-aggressive noise-reduction pass. Everything in Cast works this way. It is why Restore exists on every cut. ## Related - [Can I undo a cleanup — even long after I made it?](https://mubert.com/tools/cast/docs/undo-cleanup) - [What is the difference between Delete and Ignore?](https://mubert.com/tools/cast/docs/delete-vs-ignore) --- Source: https://mubert.com/tools/cast/docs/what-is-non-destructive · Cast docs # What is speaker diarization? Diarization is a machine working out who is speaking when, from the audio alone, and splitting the transcript by speaker automatically. Cast does not do this — you label speakers yourself, which takes a few clicks and never guesses wrong. ![Two microphones give two separable tracks and nothing has to be guessed. One shared microphone gives a single blended waveform that no editor can reliably pull apart — which is why Cast asks you to label speakers rather than guess.](https://mubert.com/tools/cast/docs-media/diagrams/what-is-diarization.svg) It is worth knowing the word because tools advertise it, and worth knowing its limits: diarization degrades badly on people talking over each other and on two voices captured by one microphone, which is exactly when you most want it to work. ## Related - [How do I separate two speakers in one audio track?](https://mubert.com/tools/cast/docs/separate-speakers) --- Source: https://mubert.com/tools/cast/docs/what-is-diarization · Cast docs # What is the difference between MP3 and WAV? WAV is the full, uncompressed audio — big files, nothing thrown away. MP3 is compressed by discarding detail people are unlikely to hear — much smaller files, and what you actually publish. A higher bitrate (320k versus 128k) means less was discarded. ![WAV keeps everything and is roughly fifteen times larger; it is what you hand on when more work is coming. MP3 discards detail you are unlikely to hear and is what you publish, because every platform re-encodes anyway.](https://mubert.com/tools/cast/docs-media/diagrams/mp3-vs-wav.svg) Publish MP3: every podcast platform re-encodes anyway, and a 320k MP3 is transparent to almost everyone. Keep or hand over WAV when the audio has more work coming — mastering, video, an archive — because each round of MP3 compression loses a little more. ## Related - [How do I export my episode as an MP3 or WAV?](https://mubert.com/tools/cast/docs/export-audio) - [Why can't I download a WAV file?](https://mubert.com/tools/cast/docs/cant-export-wav) --- Source: https://mubert.com/tools/cast/docs/mp3-vs-wav · Cast docs # The Mubert ecosystem One engine, four ways out. Which one you want depends on what you are walking away with. # Cast, Fuse or Render — which one do I want? Pick by what you are walking away with. A finished podcast episode → Cast. A short-form video with sound → Fuse. A music track on its own → Render. Music generated inside your own product → the Mubert API. They share one generative engine, so the music is the same quality wherever you meet it; what differs is the room you sit in to use it. ![The Mubert products split by what you walk out with: Cast finishes a podcast episode, Fuse scores short-form video, and the Mubert API puts music generation inside your own app or game. They share one generative music engine.](https://mubert.com/tools/cast/docs-media/diagrams/ecosystem.svg) The engine is the constant. Every one of these generates original music that did not exist before you asked for it, and that you are cleared to use — that is the thing Mubert makes. The four products are four answers to "and then what?". Cast is the only one built around a voice. That is why it is the only one that transcribes, cuts by text, and ducks the music under the speech automatically — a music tool does not need to know where the sentence ends, and a podcast finisher cannot work without it. The dividing line people get wrong: if the music is meant to sit under your own voice, generate it in Cast rather than making it in Render and importing it. Music generated in Cast arrives in stems, ducks itself, and re-follows the mix when you cut a sentence out. A downloaded track is a flat file that knows nothing about your edits — you would be back to drawing volume automation by hand. - **Cast** — You leave with a finished podcast episode: cleaned voice, scored, licensed, at the right loudness. - **Fuse** — You leave with a short-form video that sounds finished: music, SFX, subtitles, ducking. - **Render** — You leave with the music itself — a track, a bed, a loop. - **Mubert API** — You leave with nothing. Your software generates the music, at runtime, for your users. ## Related - [What is Fuse by Mubert?](https://mubert.com/tools/cast/docs/fuse) - [What is Mubert Render?](https://mubert.com/tools/cast/docs/render) - [What is the Mubert API?](https://mubert.com/tools/cast/docs/mubert-api) - [What is Cast?](https://mubert.com/tools/cast/docs/what-is-mucast) --- Source: https://mubert.com/tools/cast/docs/which-mubert-tool · Cast docs # What is the Mubert API? The Mubert API is Mubert's generative-music API for developers: you call it with a description or a mood and it returns original, license-cleared music — for apps, games, streams, or any product that needs an endless supply of music it is allowed to use. ![The Mubert products split by what you walk out with: Cast finishes a podcast episode, Fuse scores short-form video, and the Mubert API puts music generation inside your own app or game. They share one generative music engine.](https://mubert.com/tools/cast/docs-media/diagrams/ecosystem.svg) It is a developer product, not an editor. Where Cast is a place you sit and finish an episode, the Mubert API is something you build on top of — the music generation, without the interface. Reach for it when the music needs to be produced by your own software rather than by a person: a fitness app scoring a workout, a game reacting to play, a video tool generating a bed per clip. The fastest way in is the Mubert Skill — an Agent Skill that teaches an AI coding agent (Claude Code and other skills-compatible agents) the whole API: install it with “npx skills add MubertAI/skills” and ask the agent for music in plain language — it handles registration, generation, and licensing calls for you. The skill lives at https://github.com/MubertAI/skills, with setup notes at mubert.com/api/docs#skill. Full reference and access are at mubert.com/api/docs. ## Related - [What is Fuse by Mubert?](https://mubert.com/tools/cast/docs/fuse) - [What is Cast?](https://mubert.com/tools/cast/docs/what-is-mucast) --- Source: https://mubert.com/tools/cast/docs/mubert-api · Cast docs # What is Fuse by Mubert? Fuse is Mubert's AI editor for video: it scores short-form video with generated music and sound effects, ducks the audio under speech, and adds styled subtitles. Fuse and Cast share an engine and split by deliverable — Fuse makes short-form video, Cast finishes long-form episodes. **Where:** mubert.com/tools/fuse ![The Fuse key visual: a stylized portrait with the FUSE wordmark — Mubert's editor for short-form video.](https://mubert.com/tools/cast/landing/ecosystem/card-fuse.webp) The split is by what you are making. If the output is a video — a TikTok, a Reel, a Short, a talking-head clip — Fuse is the tool. If the output is audio you want to sound like a finished episode, Cast is. They share the parts that matter: the same generative music engine, the same automatic ducking, the same subtitle generation. Try it at mubert.com/tools/fuse — the free tier needs no card. ## Related - [What is the Mubert API?](https://mubert.com/tools/cast/docs/mubert-api) - [What is Mubert Render?](https://mubert.com/tools/cast/docs/render) - [What is Cast?](https://mubert.com/tools/cast/docs/what-is-mucast) --- Source: https://mubert.com/tools/cast/docs/fuse · Cast docs # What is Mubert Render? Render is where you generate music on its own — no episode, no video, no timeline. You describe what you want, it produces original licensed tracks you can download and use. **Where:** mubert.com/render ![The Mubert Render web app: prompt-driven music generation with genres, moods and playlists.](https://mubert.com/tools/cast/landing/ecosystem/card-render.webp) The three editors split by what you walk out with. Cast finishes a long-form podcast episode. Fuse turns short audio and video into polished social content. Render just gives you the music. Use Render when the music is the deliverable — a bed for a client, a loop for a stream, a track for a project that lives outside Mubert. If the music is meant to sit under your own voice, generate it inside Cast instead: it arrives in stems, ducks itself under the speech, and re-follows the mix when you cut a sentence out — which a downloaded track cannot do. Try it at mubert.com/render. ## Related - [What is Fuse by Mubert?](https://mubert.com/tools/cast/docs/fuse) - [How do I generate music for my episode?](https://mubert.com/tools/cast/docs/generate-music) - [Is the music licensed for commercial use?](https://mubert.com/tools/cast/docs/licensing) --- Source: https://mubert.com/tools/cast/docs/render · Cast docs