LiveKit Community

I need to deeply understand how the TTS configuration chunks work

Thanks for reaching out, Gustavo. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.



The TTS configuration in LiveKit, particularly for the ElevenLabs plugin, involves the use of the `chunkLengthSchedule` property within the `TTSOptions` interface. This property is an array of numbers (`number[]`) that defines how text is chunked for text-to-speech processing. Additionally, related properties like `streamingLatency` (which controls latency in streaming) and `wordTokenizer` (a WordTokenizer instance for tokenizing words) are also part of the configuration. These settings collectively allow for detailed control over how text is processed and streamed during TTS operations.
 Sources: <https://docs.livekit.io/agents-js/interfaces/plugins_agents_plugin_elevenlabs.TTSOptions.html|TTSOptions | LiveKit Agents> | <https://docs.livekit.io/agents-js/interfaces/plugins_agents_plugin_elevenlabs.TTSOptions.html|TTSOptions | LiveKit Agents>

ok, but how can I optimize the TTS to better sintethize my tokens.... which are the best practices in terms of these param configuration?

I'm sorry, but I don't have the information you need at the moment. If you would like a member of the LiveKit team to respond, please ask your question in one of the other channels.
