Integration GuidesThird-Party Integrations
OpenReader Setup
Add Fish Audio as a Custom OpenAI-Like TTS provider in OpenReader for articles, web pages, and ebooks.
Connect Fish Audio TTS to OpenReader
OpenReader reads articles and long-form text through an OpenAI-compatible endpoint and can discover voices from a custom provider.
Add a provider
Open Settings → TTS Provider and choose Custom OpenAI-Like:
| Setting | Value |
|---|---|
| API Base | https://fishaudio.org/v1 |
| API Key | Your Fish Audio API key |
| Model | fishaudio-s21pro-flash or custom |
| Voice | Select a voice returned by the service |
OpenReader requests /v1/audio/voices for discovery and sends generation requests to /v1/audio/speech. Test a short selection before reading a long article.
Docker note
OpenReader sends TTS requests from its server. Fish Audio is a public HTTPS service, so Docker deployments still use https://fishaudio.org/v1, not localhost or host.docker.internal.
Troubleshooting
- Empty voice list: verify the API key and reopen provider settings.
- Long text stops: reduce the segment length to rule out an oversized request.
- 401/402: verify the key and API credits.