Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Added curl and ffmpeg to requirements; added 30s duration check; improved script cleanup and silent transcoding.
What This Skill Does
Transcribes Chinese audio files to text using Zhipu AI's GLM-ASR model. Accepts WAV, MP3, and several other formats, converting non-native formats via ffmpeg. Supports context prompts and custom hotwords to improve accuracy for specialized vocabulary.
The hotword and context prompt features give it an edge over generic ASR tools when transcribing content with proper names, technical terminology, or multi-segment recordings.
When to use it
- Transcribing recorded business meetings in Chinese
- Converting lecture recordings to text for note-taking
- Processing interview audio for documentation
- Transcribing multi-part conversations with continuity context
- Boosting accuracy for domain-specific terms like medical or legal jargon
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: transcribe meeting.wav with attendee names Zhang and Li
- 1Check audio file format and size against 25MB/30s limits
- 2Convert format to WAV or MP3 if needed using ffmpeg
- 3Call speech_to_text.sh with the file and hotwords '张总,李经理'
- 4Send request to Zhipu AI GLM-ASR API with ZHIPU_API_KEY
- 5Parse JSON response and extract the 'text' field
Transcribed Chinese text with improved recognition of specified names
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.