Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Provides transcription, text-to-speech, voice activity detection, feature extraction, and audio transformation via a single command-line tool. Uses OpenAI Whisper for speech-to-text and Google TTS for synthesis. File paths are validated and system directory access is blocked.
Bundles five distinct audio operations into one tool so you don't need separate scripts for transcription, TTS, VAD, feature extraction, and transformation.
When to use it
- Transcribe recorded meetings or interviews to text
- Generate spoken audio from written content
- Find speech segments in long recordings
- Extract acoustic features for audio classification
- Trim, normalize, or resample audio files in batch
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: transcribe this interview recording and find where the speakers are talking
- 1Run transcribe action on the audio file using the small Whisper model
- 2Receive text transcript with timestamped segments
- 3Run vad_segments action on the same file to detect speech intervals
- 4Return transcript text alongside speech segment timestamps
Full text transcript plus a list of speech segments with start and end times