360 Secure SkillHub
360 安全技能中心

Volcengine STT

通过火山引擎语音转文字接口把本地音频文件转写为文本，默认输出同名 .txt 文件；支持 --out、--json、--language、--prompt、--model 等参数控制输出路径、返回格式与识别提示。脚本以 sh 编写，内部使用 curl 与 jq，需配置 ARK_API_KEY 等环境变量。

reed1898
reed1898
热度 63.16
最近更新 2026-05-11 17:28:08
分类：效率依赖：代码执行 / 读写文件 / 第三方 API应用场景：数据处理 / 文件转换 / API 集成

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add volcengine-stt
Metadata
name
volcengine-stt
description
Transcribe audio to text using Volcano Engine (Volcengine/ARK) speech-to-text APIs. Use when the user wants to replace Whisper/OpenAI STT with Volcengine, transcribe Telegram/Discord voice notes via Volcengine, or build a reusable STT skill for other OpenClaw agents.

Volcengine STT

Use this skill to run speech-to-text through Volcengine.

Quick start

{baseDir}/scripts/transcribe.sh /path/to/audio.ogg

Default behavior:

  • Endpoint: ${ARK_BASE_URL:-https://ark.cn-beijing.volces.com/api/v3}/audio/transcriptions
  • Model: ${ARK_STT_MODEL:-doubao-seed-asr-1-0}
  • Auth header: Authorization: Bearer $ARK_API_KEY
  • Output file: <input>.txt

Required env

  • ARK_API_KEY (required)

Optional:

  • ARK_BASE_URL (default: https://ark.cn-beijing.volces.com/api/v3)
  • ARK_STT_MODEL (default: doubao-seed-asr-1-0)

Useful flags

# Save plain text to custom path
{baseDir}/scripts/transcribe.sh ./voice.ogg --out /tmp/voice.txt

# Force model
{baseDir}/scripts/transcribe.sh ./voice.ogg --model doubao-seed-asr-1-0

# Return raw JSON (for debugging/integration)
{baseDir}/scripts/transcribe.sh ./voice.ogg --json --out /tmp/voice.json

# Hint language/prompt when needed
{baseDir}/scripts/transcribe.sh ./voice.ogg --language zh --prompt "中英混合，保留术语"

Integration notes

  • For OpenClaw voice-message handling, call this script instead of Whisper script.
  • Keep keys in machine-local config or env, never commit secrets.
  • If your Volcengine account uses a different model name, pass --model or set ARK_STT_MODEL.