Whisper transcribe

Skill oyzh888/whisper-transcribe

GPU-accelerated audio transcription for meetings, interviews, lectures. Uses WhisperX (faster-whisper + Pyannote VAD) for the best quality-speed tradeoff. Handles Chinese-English mixed audio correctly, 30-40x faster than stock whisper. Triggers on "transcribe this audio", "转录一下", "这个录音", "会议录音", "字幕", meeting m4a/wav/mp3 files, or when given audio/video file paths.From its SKILL.md

Install

npx -y skills add oyzh888/whisper-transcribe

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

5.5 KB, ~1.4k tokens by cl100k_base, as published. Nobody here has run it

whisper-transcribe — GPU meeting transcription

Batteries-included transcription for Steve's environment. Defaults tuned for 会议录音 on H100/H200 GPUs.

When to use

User says "转录", "transcribe", "字幕", "会议录音", "听写", "记录一下这个录音"
User provides an audio/video path (.m4a, .mp3, .wav, .mp4, .m4v, .mov)
Need text from audio: meetings, podcasts, interviews, lectures, screen recordings

Quick start

# Default: WhisperX large-v3, bs=16, auto language (best for Chinese+English meetings)
~/.claude/skills/whisper-transcribe/scripts/transcribe.sh <audio_file>

# Output files written to same dir as input:
#   <input>.txt                 plain transcript
#   <input>.segments.txt        [timestamp] text lines
#   <input>.json                full WhisperX JSON

Common invocations

Use case	Command
Default (meeting, auto)	`transcribe.sh meeting.m4a`
Chinese only (faster, monolingual only)	`transcribe.sh meeting.m4a --lang zh`
English only (monolingual only)	`transcribe.sh meeting.m4a --lang en`
Fast mode (bs=32, nearly identical quality)	`transcribe.sh meeting.m4a --fast`
Quality mode (bs=8, slightly better)	`transcribe.sh meeting.m4a --quality`
Speaker diarization (who said what)	`transcribe.sh meeting.m4a --diarize`
SRT subtitles	`transcribe.sh video.mp4 --srt`
Word-level timestamps	`transcribe.sh audio.wav --words`

Performance (H100 80GB, benchmarked 2026-05-05)

Audio length	Expected wall time	Realtime factor
10 min	~7s	90x
30 min	~20s	90x
60 min	~40s	90x
2 hours	~80s	90x

Model: Whisper large-v3 fp16 via CTranslate2
VAD: Pyannote V3 (auto-splits silence, handles code-switching)
GPU util: ~50% avg SM activity, ~26% compute utilization (memory-bound, normal)
VRAM: ~10 GB at bs=16, ~15 GB at bs=32

Why this skill

Compared alternatives on 54-min Chinese-English meeting:

Method	Time	Quality	Notes
openai-whisper (stock)	18 min	❌ mixes zh/en	Baseline
faster-whisper bs=32	28s	⚠️ mixes zh/en in segments	Fastest
WhisperX bs=16 (this skill)	36s	✅ Correct zh/en segmentation	Best balance

WhisperX wins on quality: Pyannote VAD correctly segments Chinese-English code-switched speech into language-consistent chunks.

Requirements

Already installed in Steve's env (verified 2026-05-05):

whisperx (pip) — includes faster-whisper + Pyannote
ffmpeg at /opt/ffmpeg/bin/ffmpeg (for audio conversion)
CUDA-capable GPU (H100/H200/A100)
HuggingFace token cached (for Pyannote VAD model, only needed once)

If running on a fresh machine, the setup script handles everything:

~/.claude/skills/whisper-transcribe/scripts/setup.sh

Agent guidance

Default is the right answer 99% of the time. Just run:

~/.claude/skills/whisper-transcribe/scripts/transcribe.sh <audio>

Only add flags when the user explicitly asks:

--lang zh / --lang en → only when audio is truly monolingual. Forcing a single language on code-switched audio causes hallucination loops. Leave blank for mixed zh/en — auto-detect + VAD handle it correctly.
--diarize → user asks "谁说了什么", "who said what", speaker labels
--srt → user asks for subtitles
--fast → user says "快一点", "quick", "draft quality is fine"
--quality → user says "最高质量", "highest accuracy"
--words → user asks for word-level timestamps or karaoke-style output

After transcription:

For short files (<5 min), show a preview of the text to user
For long files, summarize + mention output file paths
If the audio is a meeting, offer to summarize key points / extract TODOs

Troubleshooting

Problem	Fix
`Failed to load audio`	The raw WAV may lack a header; script auto-reconverts via ffmpeg
`CUDA OOM`	Use `--fast` with lower batch (8 or 4) via `--bs 4`
Poor quality on a single language	Explicitly set `--lang zh` or `--lang en`
Missing HF token (for diarization)	Run `huggingface-cli login` once
Model download slow	First run downloads ~3 GB to `~/.cache/huggingface/`; subsequent runs are cached

Script contract

scripts/transcribe.sh behavior:

Accepts any audio/video path (m4a, mp3, wav, mp4, mov, mkv, webm)
Converts to 16kHz mono WAV via ffmpeg if needed (cached next to input)
Runs WhisperX large-v3 fp16 on CUDA
Writes 3 outputs next to input file:
- <name>.txt — plain transcript (joined segments)
- <name>.segments.txt — [NNNs] text timestamped lines
- <name>.json — full WhisperX result (includes word timings if --words)
On stderr: progress, GPU utilization sample, final timing summary

Exit codes: 0 success, 1 bad input, 2 transcription error, 3 missing dependency.

What ships with it: 6 files

13.9 KB alongside SKILL.md, 3 of them executable

scripts/

setup.shruns922 B
transcribe.pyruns8.0 KB
transcribe.shruns1.0 KB

.gitignore72 B
LICENSE1.1 KB
README.md2.8 KB

Gives 0 of the 12 instructions most video audio skills give in ~1.4k tokens

Counted across 622 of the 795 authors here whose files we hold, read 2026-08-07

Read individual rule files for detailed explanationsin 21 of 622, across 10 files
Render final videoin 13 of 622, across 6 files
Use WAV PCM 16kHz mono audio formatin 12 of 622, across 3 files
Use this skill when dealing with Remotion codein 11 of 622, across 4 files
Save generated audio to a WAV filein 11 of 622, across 4 files
Handle conversion errors gracefullyin 10 of 622, across 6 files
Add captions to videos alwaysin 10 of 622, across 4 files
Generate music from text descriptions using MusicGenin 9 of 622, across 2 files
Do not skip pipeline layersin 9 of 622, across 3 files
Do not make one tool do everythingin 9 of 622, across 3 files
Use Azure Document Intelligence for complex PDFsin 9 of 622, across 4 files
Never ask the user to paste their full API keyin 9 of 622, across 3 files

Said here and by no other author read

Run the default transcription script on the audio path
Use default flags for mixed audio
Pass language flag only for monolingual audio
Show a preview for short audio files
Summarize transcript and list output paths for long files
Offer to summarize key points for meetings

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.