agentleFS
fortunto2fortunto2/solo-factory/skills/transcribe/SKILL.md

solo-transcribe

Skill18 starsChanged 4 months agoMITView on GitHub

Transcribe audio or video to text locally — on the Apple GPU with mlx-whisper large-v3, or faster-whisper on CPU/CUDA — with the language decided per stretch of speech, .srt subtitles, word timings for editing, and optional speaker labels. Nothing is uploaded. Use when user says "расшифруй", "транскрибируй", "transcribe this", "make a transcript", "субтитры из видео", "что там говорят в записи", or hands a recording, a call, a voice memo or a folder of clips that needs its speech as text. Do NOT use for listening to a live room (use /supervox) or for cutting the footage itself (use /video-edit).

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Unknown names become celebrities: Keep a glossary per project (people, places, products) and fix the transcript once, before anything …

    line 60
  2. Distinctive:

    Write transcripts where the task lives, never into an unrelated repo.

    line 76

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

transcribe — local speech to text, GPU first

Any ffmpeg-decodable file becomes <name>.txt ([MM:SS] (lang) text), <name>.srt, and on request <name>.words.json with per-word timings, the input a video edit needs. Audio never leaves the machine; the only network step is a one-time model download.

Pick the engine

MachineScriptWhy
Apple Siliconscripts/transcribe_mlx.pylarge-v3 on the GPU. Measured: 2 min of a noisy prize announcement in 16 s, where medium on CPU returned fragments; a 38-min talk in 8.7 min (4x realtime)
NVIDIA / Linux, or speaker labels neededscripts/transcribe.pyfaster-whisper (CTranslate2): CUDA yes, Metal no, so on a Mac it is CPU-only however fast the Mac
uv run scripts/transcribe_mlx.py rec.m4a                      # → rec.txt, rec.srt
uv run scripts/transcribe_mlx.py clip.mp4 --words             # + rec.words.json for editing
uv run scripts/transcribe_mlx.py call.m4a --languages en ru   # languages that can occur
uv run scripts/transcribe.py rec.m4a --model large-v3         # CPU/CUDA path
uv run scripts/transcribe.py call.m4a --multilingual          # per-segment language on that path
uv run --with pyannote.audio scripts/transcribe.py talk.wav --diarize --speakers Alice Bob

Both scripts declare their dependencies in a PEP 723 header, so uv run needs no venv. Diarization needs HUGGINGFACE_TOKEN with the pyannote model terms accepted; without it the transcript comes back unlabelled, and the script says so.

Long files (10+ min): run in the background and wait for ✅ done.

Gotchas

  • Whisper picks one language from the first 30 s and applies it to the whole file. A call that opens in English and moves to Russian comes out as fluent, confident English, or the second language simply vanishes (measured: a Russian+English file detected as ru returned only the Russian words). transcribe_mlx.py splits on silences (Silero VAD) and detects per chunk; on the CPU path pass --multilingual.
  • A chunk detected as a language not in --languages (Russian read as Ukrainian, a laugh as Welsh) is re-run in the first non-English listed language instead of keeping the guess.
  • Accented English is detected as the speaker's native language and comes back translated. A Russian speaker asking a question in English at a talk, with --languages ru en, returned a fluent Russian sentence the speaker never said. For a recording known to be one language, force it (--languages en) and keep the multi-language run only where the language really switches.
  • A quiet distant voice can vanish under the VAD. At a panel filmed from the audience, a speaker ~9 dB below the person next to the camera (−31 vs −22.5 dB mean) lost two minutes of her answer: the transcript jumps from 02:53 to 04:26 with no gap marker. Check long gaps in the timestamps; a second recording (a laptop transcriber in the room) or a normalised re-run fills them.
  • Crowd noise hallucinates. 25 minutes of a bar came back as lang=fr with 454 invented sentences. Treat such a file as no speech; if one voice in it matters, re-run with the language forced.
  • Unknown names become celebrities. Keep a glossary per project (people, places, products) and fix the transcript once, before anything quotes it.
  • Screen recordings on macOS often have no audio track. ffprobe first.
  • Parakeet and other English-only models are fast but drop every non-English clip; check the language mix before choosing them for a folder.

Finding recordings on a Mac

  • Zoom: ~/Documents/Zoom/<date time …>/audioNNN.m4a (folder name = start, file mtime ≈ end).
  • Voice Memos: names and files are in ~/Library/Group Containers/group.com.apple.VoiceMemos.shared/Recordings/CloudRecordings.db (ZCLOUDRECORDING: ZCUSTOMLABEL, ZPATH, ZDATE + 978307200 = unix start, ZDURATION).
  • Long video on a slow card: pull only the audio, ffmpeg -i in.mp4 -vn -c:a copy out.m4a.

Privacy

Write transcripts where the task lives, never into an unrelated repo. Personal recordings belong in the user's own private store, not next to project code.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.