Retake
yyert123/retake
Fix a voiceover mistake without re-recording: re-voice one sentence in your voice and splice it in.
- tools
- 2
- auth
- Secret needed
- transport
- STDIO
- env var
- 1
- last updated
- Nov 8, 2026
Tools
2 toolsPublic tool metadata for what this MCP can expose to an agent.
transcribe_takeTranscribes a recording into numbered sentences with timestamps.
Transcribes a recording into numbered sentences with timestamps.
No parameter schema in public metadata yet.
fix_lineRe-voices a sentence in a recording, replacing old words with new words or a line number with new text.
Re-voices a sentence in a recording, replacing old words with new words or a line number with new text.
No parameter schema in public metadata yet.
Sponsors
Promote your product hereReach 125k+ visitors monthlyPromote your product hereReach 125k+ visitors monthly
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any language
Agent CaffeineSponsoredClose your lid, come back to finished workSnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthlyWaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthlyknobsSponsoredSee your site on any phone, any day, in any languageConfiguration
MIRAGEA_API_KEY*secretMiragea API key (https://miragea.ai), used for transcription and voice
Readme
Retake: fix a voiceover mistake in your own voice (open-source Overdub alternative)
Said "Wednesday" instead of "Thursday"? Don't re-record. Retake re-voices only the sentence you changed, in the same voice, and splices it back into your recording with matched loudness, tone (EQ) and room tone. Works with any editor (Premiere, Resolve, Final Cut, Audacity, CapCut) because it simply gives you a fixed WAV plus edit markers. CLI for humans, MCP server for AI agents. Voice engine: Miragea.
pip install retake-vo
export MIRAGEA_API_KEY=your_key # https://miragea.ai
retake transcribe take.wav # numbered sentences + timestamps (cached, pay once)
retake fix take.wav --replace "Wednesday" --with "Thursday" --voice YOUR_VOICE_ID
# -> take.fixed.wav, take.fixed.edits.json, take.fixed.markers.csv
Details
- Repository
- github.com/yyert123/retake
- Website
- huggingface.co/spaces/headvox/retake
- Package
retake-vo- Transport
- STDIO
- Auth
- Required
- Registry
- MCP registry, active