LiveKit supports both Speechify and ElevenLabs for TTS voice cloning in live calls, and plugins are available for seamless integration. Here’s how you can set them up: Speechify Integration: • Installation: Run the following command to install the Speechify plugin:

pip install "livekit-agents[speechify]~=1.0"

• Authentication: Set your

SPEECHIFY_API_KEY

in the

.env

file. • Usage: Use the following code snippet to configure Speechify in your session:

from livekit.plugins import speechify

session = AgentSession(
    tts=speechify.TTS(
        model="simba-english",
        voice_id="jack",
    )
    # ... llm, stt, etc.
)

ElevenLabs Integration: • Installation: Run the following command to install the ElevenLabs plugin:

pip install "livekit-agents[elevenlabs]~=1.0"

• Authentication: Set your

ELEVEN_API_KEY

in the

.env

file. • Usage: Use the following code snippet to configure ElevenLabs in your session:

from livekit.plugins import elevenlabs

session = AgentSession(
    tts=elevenlabs.TTS(
        voice_id="ODq5zmih8GrVes37Dizd",
        model="eleven_multilingual_v2"
    )
    # ... llm, stt, etc.
)

Comparison: • Speechify: Offers ultra-low latency, high-quality voice cloning with support for SSML for custom pronunciation. • ElevenLabs: Provides thousands of human-like voices across multiple languages and robust voice cloning capabilities. Both options are suitable for live calls, and the choice depends on your specific requirements for voice quality, language support, and customization. Sources: Speechify TTS integration guide | LiveKit Docs | ElevenLabs TTS integration guide | LiveKit Docs | ElevenLabs TTS integration guide | LiveKit Docs | Text-to-speech (TTS) integrations | LiveKit Docs