| name | audio_transcriber |
| description | **Problem Category**: Audio file transcription using OpenAI Whisper |
audio_transcriber
Description
Problem Category: Audio file transcription using OpenAI Whisper
Applicable Questions: Transcribing audio files (mp3, wav, etc.) to text, especially Chinese audio
Key Features:
- Uses OpenAI Whisper model for transcription
- Supports Chinese language transcription
- Falls back to Google Speech Recognition if Whisper fails
- Auto-installs required dependencies
- Extracts file path from natural language query via LLM
Skills Used: execute_shell_command
Reasoning Pattern: 1) Parse query for file path 2) Verify file exists 3) Install whisper 4) Transcribe with whisper (base model, zh language) 5) Fallback to SpeechRecognition if needed 6) Return transcribed text
Input Format: Natural language query containing an audio file path (e.g., 'Transcribe the audio file at /path/to/file.mp3')
Output Format: {answer: transcribed text, summary: process description}
Skills Used
playwright, google_search, execute_shell_command
Usage
Entry file: audio_transcriber.py
Query type: Pass a focused sub-question as the query.
How to call:
<action>run_subagent</action>
<params>{"skill_name": "audio_transcriber", "query": "<your focused sub-question>"}</params>