OverviewHistoryStatsSecurity
npx skills add ...
Documentation
SKILL.md
npx skills add marswaveai/skills --skill asr
Transcribe audio files to text using local speech recognition. Triggers on: "转录", "transcribe", "语音转文字", "ASR", "识别音频", "把这段音频转成文字".
npx skills add marswaveai/skills --skill asr
/tts)/podcast or /explainer)Transcribe audio files to text using coli asr, which runs fully offline via local
speech recognition models. No API key required. Supports Chinese, English, Japanese,
Korean, and Cantonese (sensevoice model) or English-only (whisper model).
Run coli asr --help for current CLI options and supported flags.
shared/config-pattern.md before any interactionshared/cli-patterns.md for interaction patternsBefore config setup, silently check the environment:
| Issue | Action |
|---|---|
coli not found | Block. Tell user to run npm install -g @marswave/coli first |
ffmpeg not found | Warn (WAV files still work). Suggest brew install ffmpeg / sudo apt install ffmpeg |
| Models not downloaded | Inform user: first transcription will auto-download models (~60MB) to ~/.coli/models/ |
If coli is missing, stop here and do not proceed.
Follow shared/config-pattern.md Step 0 (Zero-Question Boot).
If file doesn't exist — silently create with defaults and proceed:
Do NOT ask any setup questions. Proceed directly to the Interaction Flow with sensible defaults (sensevoice model, polish enabled).
If file exists — read config silently and proceed:
Only run when the user explicitly asks to reconfigure. Display current settings:
Ask in order:
model: "默认使用哪个语音识别模型？"
polish: "转录后由 AI 润色文本？（修正标点、去语气词、提升可读性）"
polish: truepolish: falseSave all answers at once after collecting them.
If the user hasn't provided a file path, ask:
"请提供要转录的音频文件路径。"
Verify the file exists before proceeding.
Run coli asr with JSON output (to get metadata):
On first run, coli will automatically download the required model. This may take a
moment — inform the user if models haven't been downloaded yet.
Parse the JSON result to extract text, lang, emotion, event, duration.
If polish is true, take the raw text from the transcription result and rewrite
it to fix punctuation, remove filler words, and improve readability. Preserve the
original meaning and speaker intent. Do not summarize or paraphrase.
Display the transcript directly in the conversation:
If polished, show the polished version with a note that it was AI-refined. Offer to show the raw original on request.
After presenting the result, ask:
If yes, write {audio-filename}-transcript.md to the current working directory
(where the user is running Claude Code). The file should contain the transcript text
(polished version if polish was enabled), with a front-matter header:
"帮我转录这个文件 meeting.m4a"
coli asr -j --model sensevoice "meeting.m4a""transcribe interview.wav, no polish"
coli asr -j --model sensevoice "interview.wav"mkdir -p ".listenhub/asr"
echo '{"model":"sensevoice","polish":true}' > ".listenhub/asr/config.json"
CONFIG_PATH=".listenhub/asr/config.json"
CONFIG=$(cat "$CONFIG_PATH")CONFIG_PATH=".listenhub/asr/config.json"
[ ! -f "$CONFIG_PATH" ] && CONFIG_PATH="$HOME/.listenhub/asr/config.json"
CONFIG=$(cat "$CONFIG_PATH")当前配置 (asr)：
模型：sensevoice / whisper-tiny.en
润色：开启 / 关闭准备转录：
文件：{filename}
模型：{model}
润色：{是 / 否}
继续？coli asr -j --model {model} "{file}"转录完成
{transcript text}
─────────────────
语言：{lang} · 情绪：{emotion} · 时长：{duration}sQuestion: "保存为 Markdown 文件到当前目录？"
Options:
- "是" — save to current directory
- "否" — done---
source: {original audio filename}
date: {YYYY-MM-DD}
model: {model used}
duration: {duration}s
lang: {detected language}
---
{transcript text}