Integration GuidesThird-Party Integrations
LiveKit Agents Setup
Configure the Kitta Audio Base URL, model, and voice in the LiveKit Agents OpenAI TTS plugin.
Connect Kitta Audio TTS to LiveKit Agents
The LiveKit Agents OpenAI plugin can override the TTS base_url, API key, model, and voice for realtime assistants, support agents, and telephony.
Python
import os
from livekit.agents import AgentSession
from livekit.plugins import openai
session = AgentSession(
tts=openai.TTS(
api_key=os.environ["FISH_AUDIO_API_KEY"],
base_url="https://fishaudio.org/v1",
model="fishaudio-s21pro-flash",
voice="00a1b221-6137-4b73-ad62-b0cbce134167",
),
)Install the plugin with uv add "livekit-agents[openai]~=1.5".
Node.js
import { AgentSession } from '@livekit/agents';
import * as openai from '@livekit/agents-plugin-openai';
const session = new AgentSession({
tts: new openai.TTS({
apiKey: process.env.FISH_AUDIO_API_KEY!,
baseURL: 'https://fishaudio.org/v1',
model: 'fishaudio-s21pro-flash',
voice: '00a1b221-6137-4b73-ad62-b0cbce134167',
}),
});Test a short sentence end to end before adding SIP, WebRTC, or a frontend. Keep the API key on the agent server.
Troubleshooting
- Voice type is rejected: current LiveKit versions accept string Voice IDs; upgrade an older plugin.
- Request succeeds but no audio plays: inspect the audio pipeline sample rate and output device.
- 401/402: verify the environment variable and API credits.