Here is a complete example for integrating Speechify TTS with LiveKit: 1. First, install the required package:

pip install "livekit-agents[speechify]~=1.2"

2. Next, create a

.env

file in your project root and add your Speechify API key:

SPEECHIFY_API_KEY=your_api_key_here

3. Then, use the following Python code for the integration:

import asyncio
import logging
from dotenv import load_dotenv
from livekit import rtc
from livekit.agents import AutoSubscribe, JobContext, WorkerOptions, cli
from livekit.agents.tts import TTSCapabilities
from livekit.plugins import speechify

load_dotenv()

logger = logging.getLogger("speechify-tts-demo")
logger.setLevel(<http://logging.INFO|logging.INFO>)

async def entrypoint(ctx: JobContext):
    try:
        await ctx.connect(auto_subscribe=AutoSubscribe.SUBSCRIBE_NONE)
        participant = await ctx.wait_for_participant()
        <http://logger.info|logger.info>("Connected to LiveKit room and participant joined")

        tts = speechify.TTS(voice_id="jack", model="simba-english", language="en", encoding="mp3_24000", loudness_normalization=True, text_normalization=True)
        assert isinstance(tts.capabilities, TTSCapabilities)
        <http://logger.info|logger.info>(f"TTS initialized with sample rate: {tts.sample_rate}")

        source = rtc.AudioSource(tts.sample_rate, tts.num_channels)
        track = rtc.LocalAudioTrack.create_audio_track("tts-audio", source)
        options = rtc.TrackPublishOptions()
        options.source = rtc.TrackSource.SOURCE_MICROPHONE

        publication = await ctx.room.local_participant.publish_track(track, options)
        await publication.wait_for_subscription()
        <http://logger.info|logger.info>("Audio track published successfully")

        text = "Hello! I'm using Speechify TTS with LiveKit Agents to convert this text into natural-sounding speech in real time."
        async for output in tts.synthesize(text):
            await source.capture_frame(output.frame)

        <http://logger.info|logger.info>("TTS synthesis completed successfully")

    except Exception as e:
        logger.error("An error occurred in the TTS agent: %s", str(e), exc_info=True)
        raise
    finally:
        if 'tts' in locals():
            await tts.aclose()

if __name__ == "__main__":
    cli.run_app(WorkerOptions(entrypoint_fnc=entrypoint))

If you'd like, you can also explore listing available voices programmatically or using SSML for more advanced speech control. Please feel free to reach out if you have any questions or need a hand with anything else—I’m here to help! Sources: Speechify TTS integration guide | LiveKit Docs | livekit-plugins/livekit-plugins-speechify/pyproject.toml