# Remove retakes from video and audio: bad take remover

Cut false starts and repeated lines automatically while keeping your final take intact.

Powered by [Text-based editor](https://audo.ai/text-based-editor.md) — cuts earlier tries of a repeated sentence in the transcript.

## How it works

1. Add a video or audio recording, leave Spoken language on automatic detection (or pick one), and switch on Also cut filler words if needed.
2. Press Remove retakes to transcribe speech and open Audo's editor in Transcript mode with earlier takes struck through.
3. Click any struck-through line to restore an earlier take, then open Export for MP4, WAV, MP3, or a free video editor timeline.

## Recordings where a retake remover saves rough-cut hours

Deliver lines two or three times on camera without stopping record — flubbed attempts are trimmed while your clean take stays.

### YouTube talking heads: Auto cut video rough cuts without scrubbing waveforms

Solo YouTube creators often restart sentences mid-thought until the pacing lands. Use this retake remover to auto cut video takes in one pass, and compare rough-cut options in our [Gling alternative](https://audo.ai/guides/gling-alternative.md) guide.

### Courses & tutorials: Cut false starts from screen recordings and lessons

Stumble over a technical step in an OBS, Loom, or webcam lesson? Say “let me start over” or repeat the sentence cleanly, and picture and sound are trimmed together in the quiet gap between words.

### Podcasts & voiceovers: Clean solo intros, ad reads, and narration pickups

Sponsor reads, host intros, and audiobook pickups routinely stack three tries in a row. Combine this bad take remover with [Remove filler words](https://audo.ai/remove-filler-words.md) and [Remove silence](https://audo.ai/silence-remover.md) for smooth, natural pacing.

### Premiere, Resolve & Final Cut: Export non-destructive retake cuts to your desktop timeline

Keep color grading, B-roll, and music mixing inside Premiere Pro, DaVinci Resolve, or Final Cut Pro by downloading a free XML or FCPXML timeline from Export — see our [Descript alternative](https://audo.ai/guides/descript-alternative.md) guide.

## How Audo detects repeated takes and spoken reset cues

Audo opens your recording in Transcript mode with Retakes enabled, scanning timed speech across audio (WAV, MP3, M4A, AAC, FLAC, OGG, Opus, WebM, AIFF) and video (MP4, MOV, MKV, WebM, AVI) files up to 10 GB and 10 hours signed in. Whenever the same speaker repeats a sentence within 30 seconds — matching at least 60% of its words in order from the same opening — earlier tries are cut and the final take stays.

Speakers can also reset a line aloud using cues like “let me say that again”, “let me start over”, “scratch that”, “take two”, or “one more time” in 7 cue languages: English, German, French, Spanish, Italian, Dutch, and Portuguese. Short phrases of three words or fewer stay untouched unless followed by a cue, and cuts land in the quiet valley between words.

Every cut remains visible and struck through in the transcript so a single click restores an earlier delivery. Inside the editor's Cut panel, pair Retakes with [Remove filler words](https://audo.ai/remove-filler-words.md) and Shorten pauses, or trim custom passages by hand in the [Text-based editor](https://audo.ai/text-based-editor.md).

Rough-cut controls inside Audo's editor Cut panel

| Editor control | What Audo cuts or exports | How you review it |
|---|---|---|
| Retakes | Sentences repeated within 30 s (60%+ words in order) or followed by a reset cue in 7 languages | Struck through in transcript; one click restores any take |
| Remove filler words | Ums, uhs, and equivalents across 8 languages plus adjacent word repeats | Switch on Also cut filler words or toggle inside Cut |
| Shorten pauses | Long dead air tightened down to a natural breath in the browser | Adjust pause threshold and kept duration live |
| Export | Rendered MP4, WAV, or MP3 files, or timelines for Premiere Pro, DaVinci Resolve, and Final Cut Pro | Timeline files reference your original camera footage |

Tool: Transcribe (`transcribe_audio`). The same tool's other pages: https://audo.ai/transcribe.md, https://audo.ai/speaker-diarization.md, https://audo.ai/podcast-transcription.md, https://audo.ai/interview-transcription.md, https://audo.ai/srt-file-generator.md, https://audo.ai/text-based-editor.md, https://audo.ai/m4a-to-text.md. Price: 3 credits a minute — $0.03/min ($1.80/hr), pay-as-you-go with no monthly subscription — of input for whole files, charged only when a job succeeds.

## Free API

```bash
curl -F file=@take.mp4 -F verbatim=true https://audo.ai/api/try/transcribe -D response.headers -o response.body
```

Read response.headers and response.body before naming media. Handle 202, 200 JSON and errors with the complete response client at https://audo.ai/docs/api?utm_source=markdown&utm_medium=agent#responses.

Free for the first minute of each file, 3 times a day per person, with no key or account. Uploads up to 10 GB, 2 at once from one address. The response is the result itself; a job that runs longer than about 30 seconds returns 202 with a link to check on it. With an API key (`-H "Authorization: Bearer $AUDO_API_KEY"`), omitting free runs the full job with credits if the balance covers it, otherwise the free allowance if quota remains. Set free=true to prohibit spending; for paid runs use free=false with an approved positive max_credits. A free result cut short comes with finishUrl (and the Audo-Finish-Url header when the response is the file itself): give it to the person, who pays there (from $5, no sign-up), and the whole file runs; keep checking the status link, whose finishedBy names the whole-file job and its own status link. Details: https://audo.ai/docs/api.md

## Options

Form fields with the same names as the MCP tool's inputs:

- `language`: The spoken language, as a code such as en or ja. Default: detected.
- `verbatim`: Keeps filler words such as um and uh, so they can be cut later. Default true.
- `speaker_labels`: Labels who speaks when: Speaker 1, Speaker 2, and so on, in the order they first speak, up to 8 people. No names. Default true.
- `format`: what the free API returns: json (default), srt, vtt, or txt.

## Speakers

Transcription includes automatic speaker diarization and labels who speaks when, unless `speaker_labels` is false: Speaker 1, Speaker 2, and so on, numbered in the order people first speak, without real names. It finds at most 8 speakers.

- JSON: every word and sentence has `speaker` (a number from 1), and the transcript has `speaker_count`. A sentence never spans two speakers.
- TXT: with two or more speakers, one paragraph per turn, starting "Speaker 1: ".
- VTT: with two or more speakers, each cue names its speaker in a voice span, such as `<v Speaker 1>`.
- SRT: no names, so it stays clean for YouTube.
- No caption spans two speakers.
- The job's result has `speaker_labels` (`on`, `off`, or `failed`) and `speaker_count`.
- Reading the transcript: MCP `get_transcript` gives `speaker_count` and each sentence's `speaker`; the REST transcript endpoint gives `speakerCount` and `speaker`.

If labeling fails, the job still succeeds with every word, just without speakers: `speaker_labels` is `failed`, and the result has a `warning`.

## Languages

60 languages, detected automatically. To name one, set `language` to its code: `af` Afrikaans, `ar` Arabic, `hy` Armenian, `as` Assamese, `az` Azerbaijani, `bn` Bengali, `bs` Bosnian, `bg` Bulgarian, `yue` Cantonese, `ca` Catalan, `zh` Chinese (Simplified), `cs` Czech, `da` Danish, `nl` Dutch, `en` English, `et` Estonian, `fil` Filipino, `fi` Finnish, `fr` French, `gl` Galician, `de` German, `el` Greek, `gu` Gujarati, `he` Hebrew, `hi` Hindi, `hu` Hungarian, `is` Icelandic, `id` Indonesian, `it` Italian, `ja` Japanese, `kn` Kannada, `kk` Kazakh, `ko` Korean, `lv` Latvian, `lt` Lithuanian, `mk` Macedonian, `ms` Malay, `ml` Malayalam, `mr` Marathi, `ne` Nepali, `nb` Norwegian Bokmål, `or` Odia, `fa` Persian, `pl` Polish, `pt` Portuguese, `pa` Punjabi, `ro` Romanian, `ru` Russian, `sk` Slovak, `sl` Slovenian, `es` Spanish, `sw` Swahili, `sv` Swedish, `ta` Tamil, `te` Telugu, `th` Thai, `tr` Turkish, `uk` Ukrainian, `ur` Urdu, `vi` Vietnamese.

## From an assistant (MCP)

Add https://audo.ai/mcp and sign in (setup for each assistant: https://audo.ai/setup.md). The tool is `transcribe_audio`:

> Transcribes speech in an audio or video file, with a start and end time for every word and who speaks when (Speaker 1, Speaker 2, and so on, in the order they first speak). Use it when the person wants text or captions, or wants to cut by words. Not for lining up a script they already have (align_script). It costs 3 credits a minute, charged only if the job succeeds. Returns TXT, SRT, VTT, and JSON files, which get_transcript reads, or job_id for get_job if it takes longer.

Example: "Use Audo to transcribe take.mp4 with word timings so I can cut the retakes."

## Questions

**Is it free to remove retakes from video and audio?** Yes — Audo transcribes the first minute of each recording free, 3 times a day, with no sign-up, and whole recordings cost $0.03 a minute ($1.80 an hour) from $5 with no subscription. Finding and cutting retakes in the editor (plus Premiere Pro, DaVinci Resolve, or Final Cut Pro timeline export) is free; exporting rendered audio or video uses Cut's rate ($0.01 a minute, the first hour free 10 times a day) — [see pricing](https://audo.ai/pricing.md).

**What counts as a retake, and which take does Audo keep?** Audo marks a retake when the speaker repeats at least 60% of a sentence's words in order from the same start within 30 seconds, or follows it with a cue like “let me say that again”, “let me start over”, “scratch that”, “take two”, or “one more time”. Earlier attempts and cues are cut, the final take stays, and clicking any struck-through sentence restores it.

**Which languages support retake matching and spoken reset cues?** On Audo, repeated-sentence matching works across all 60 languages in [Transcribe audio](https://audo.ai/transcribe.md), while spoken reset cues are built in for 7 languages: English, German, French, Spanish, Italian, Dutch, and Portuguese.

**Can I export retake cuts to Premiere Pro, DaVinci Resolve, or Final Cut Pro?** Yes. Open Export in Audo's editor and choose Premiere Pro, DaVinci Resolve, or Final Cut Pro to download a free XML or FCPXML timeline that points directly at your original file.

**Can I cut filler words and shorten pauses at the same time?** Yes. On Audo, turn on Also cut filler words before pressing Remove retakes, then toggle Shorten pauses inside the editor's Cut panel. For standalone cleanup tools, compare [Cleanvoice alternatives](https://audo.ai/guides/cleanvoice-alternatives.md).

**What if no retakes are found, or I repeat a short two-word phrase?** Audo ignores sentences of three words or fewer unless followed by a spoken reset cue, so deliberate emphasis stays intact. When zero retakes are detected, nothing is cut and you can edit words manually.

**Can an AI assistant or script get word timings for retake cutting?** Yes. [Connect Audo over MCP](https://audo.ai/setup.md) or call the [transcription API](https://audo.ai/docs/transcription-api.md) with verbatim word timings from ChatGPT, Claude, or your own pipeline, then review retake cuts visually in Audo's browser editor.

## More

- Other ways to make captions: [Transcribe audio](https://audo.ai/transcribe.md), [Text to SRT](https://audo.ai/text-to-srt.md), [Auto captions](https://audo.ai/auto-captions.md), [YouTube chapter generator](https://audo.ai/youtube-chapter-generator.md), [Text-based editor](https://audo.ai/text-based-editor.md), [Speaker diarization](https://audo.ai/speaker-diarization.md), [M4A to text](https://audo.ai/m4a-to-text.md), [Podcast transcription](https://audo.ai/podcast-transcription.md), [Interview transcription](https://audo.ai/interview-transcription.md), [SRT file generator](https://audo.ai/srt-file-generator.md), [Podcast show notes generator](https://audo.ai/podcast-show-notes-generator.md), [Meeting shortener](https://audo.ai/meeting-shortener.md)
- Related guides: [Descript alternative for transcripts and text edits](https://audo.ai/guides/descript-alternative.md), [Gling alternative for cutting bad takes and pauses](https://audo.ai/guides/gling-alternative.md), [Best AI transcription tools](https://audo.ai/guides/best-ai-transcription-tools.md)
- [All tools](https://audo.ai/all-tools.md)
- [Prices and free limits](https://audo.ai/pricing.md)
- [Set up your assistant](https://audo.ai/setup.md)
- Every tool: https://audo.ai/llms.txt
- OpenAPI: https://audo.ai/api/openapi.json
- This page for people: https://audo.ai/remove-retakes?utm_source=markdown&utm_medium=agent
