Kokoro FastAPI
Turns agent replies into speech through an OpenAI-compatible endpoint, with voice mixing, multiple languages, and mp3, wav, opus, or flac output.
What it provides
Connect an HTTPS inference API. Model weights stay on CPU/GPU infrastructure and are never installed inside the 4 GB bot.
Connection requirements
Run the Kokoro FastAPI container outside the bot (it listens on port 8880) and publish it through an authenticated HTTPS reverse proxy, then keep Bearer selected and paste the proxy token below. The upstream image ships no authentication of its own, so never expose it directly. Use post-save so generated audio lands in a file instead of the chat transcript.
- Protocol
- OpenAI-compatible REST
- Endpoint
https://kokoro.example.com- Authentication
- bearer
Where to add it
Open Dashboard Tools, select a running OpenClaw or Hermes instance, then use the card’s Install or Connect action. External services may require an HTTPS endpoint and credentials; the dashboard shows those fields before anything is saved.