Fish Audio Docs
Integration GuidesThird-Party Integrations
Back to site

Dify Setup

Use Fish Audio TTS with the Dify Podcast Studio plugin to generate single-speaker or two-host podcast audio.

Connect Fish Audio TTS to Dify

Dify's Podcast Studio plugin supports a custom OpenAI TTS Base URL. Convert workflow-generated scripts into Fish Audio podcasts or two-speaker conversations.

Install and authorize

  1. Install Podcast Studio from Dify Marketplace.
  2. Open plugin authorization and choose OpenAI TTS.
  3. Enter the API key and custom Base URL.
  4. Save and run the connection test.
SettingValue
Base URLhttps://fishaudio.org/v1
API KeyYour Fish Audio API key
Modelfishaudio-s21pro-flash
Host 1 Voice00a1b221-6137-4b73-ad62-b0cbce134167
Host 2 VoiceAnother Fish Audio Voice ID

Build a podcast workflow

Recommended flow:

Topic input → LLM creates a two-host script → Podcast Studio → Audio file

Prompt the LLM to alternate speakers line by line, then assign one Voice ID to each host. Both can use the system test voice for the first run.

Troubleshooting

  • No Base URL field: install a Podcast Studio version that supports a custom OpenAI endpoint.
  • Speakers are reversed: ensure the script alternates speakers one line at a time.
  • Invalid model or voice: enter Fish Audio model and Voice IDs manually.
  • 401/402: verify plugin credentials and API credits.

PocketTavern Setup

Configure Fish Audio TTS in the PocketTavern Android app and assign a different voice to each character.

Open Notebook Setup

Add Fish Audio OpenAI-compatible TTS credentials and a speech model to Open Notebook for podcasts and audio summaries.

On this page

Connect Fish Audio TTS to DifyInstall and authorizeBuild a podcast workflowTroubleshootingRelated documentation