vahnxu

Doubao Asr — Development skill for Claude Code

Development community

Agent Skill: Transcribe audio files via ByteDance Volcengine Seed-ASR 2.0 (豆包录音文件识别模型2.0).

How to install Doubao Asr

This entry records only its repository, not the path inside it, so there is no exact command to give. Open vahnxu/doubao-asr and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Doubao Asr does

Agent Skill: Transcribe audio files via ByteDance Volcengine Seed-ASR 2.0 (豆包录音文件识别模型2.0). Best-in-class Chinese speech recognition with step-by-step setup guide.

Alternatives in Development

Transcribe — Transcribe audio files to text with optional speaker diarization 14.6k ★
GenericAgent — Self-evolving agent: grows skill tree from 3.3K-line seed, achieving full system control with 6x less token co 6.1k ★
Stt System — Transcribe this voice message accurately 198 ★

README

doubao-asr

Agent Skill for transcribing audio files via ByteDance Volcengine **Seed-ASR 2.0** (豆包录音文件识别模型2.0)

Best-in-class Chinese speech recognition — Mandarin, Cantonese, Sichuan dialect, and 13+ languages. With speaker diarization, SRT subtitle export, and three recognition tiers (standard / express / offpeak), up to 5 hours / 512MB per file.

中文语音识别准确率业界领先——支持普通话、粤语、四川话等方言及 13+ 种语言。自带说话人分离、SRT 字幕导出、三档识别版本（标准/极速/闲时），单文件最长 5 小时 / 512MB。

Why this skill?

Setting up Volcengine's Doubao Audio File Recognition 2.0 (豆包录音文件识别模型2.0, recorded audio → text) from scratch involves **4 environment variables across 3 different console pages** (Speech console, IAM, TOS). Without guidance, this typically takes 1-2 hours of doc-hunting and trial-and-error.

This skill provides **step-by-step bilingual setup instructions** (中英双语) baked into `SKILL.md`, so your AI agent can walk you through the entire process in ~10 minutes.

从零配置火山引擎豆包录音文件识别模型2.0（录音转文字）涉及 **3 个不同控制台页面的 4 个环境变量**。没有引导的话通常需要 1-2 小时翻文档踩坑。本 skill 在 `SKILL.md` 中内置了**中英双语分步引导**，AI agent 可以在约 10 分钟内带你完成全部配置。

Install

Claude Code

claude /install-skill https://github.com/vahnxu/doubao-asr

Or manually copy to `~/.claude/skills/doubao-asr/`.

OpenClaw / ClawHub

Available on [ClawHub](https://clawhub.com) as `doubao-asr`.

Manual

git clone https://github.com/vahnxu/doubao-asr.git
cd doubao-asr
pip install requests
python3 scripts/transcribe.py /path/to/audio.m4a

Quick start

# Basic transcription
python3 scripts/transcribe.py /path/to/audio.m4a

# Save to file
python3 scripts/transcribe.py /path/to/audio.m4a --out /tmp/transcript.txt

# JSON output
python3 scripts/transcribe.py /path/to/audio.m4a --json --out /tmp/result.json

# Direct URL (skip upload)
python3 scripts/transcribe.py https://example.com/audio.mp3

# Disable speaker diarization
python3 scripts/transcribe.py /path/to/audio.m4a --no-speakers

# Export SRT subtitles (from per-utterance timestamps)