Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Provides access to text-to-speech, speech-to-text, voice conversion, and audio processing through the EachLabs Predictions API. Supports ElevenLabs TTS and transcription models, Whisper-based transcription with speaker diarization, RVC voice conversion, and audio/video utilities. Authentication requires a single API key.
Consolidates over 30 voice and audio models behind a single API key and uniform prediction flow, removing the need to manage separate credentials and SDKs for ElevenLabs, Whisper, RVC, and Stable Audio.
When to use it
- Transcribe a recorded meeting and label each speaker
- Convert a script to spoken audio using a custom voice
- Dub a video into a different voice or language
- Generate sound effects from a text description
- Merge a narration audio track with a silent video file
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: transcribe this meeting recording and identify who is speaking
- 1Check the whisper-diarization model schema via GET /v1/model?slug=whisper-diarization
- 2POST to /v1/prediction with the audio file URL, num_speakers, and language set to en
- 3Poll GET /v1/prediction/{id} until status is success
- 4Extract the grouped transcript segments with speaker labels from the response
- 5Return the labeled transcript to the user
A transcript with each segment attributed to a speaker, e.g. Speaker 1: ... Speaker 2: ...
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.