LiveKit offers several starter code options for implementing a voice assistant that can be used during meeting calls. Here are the main options: 1. Python Voice Assistant Starter The Python Voice Agent starter project provides a simple implementation for creating a voice assistant. It includes features like Voice Activity Detection (VAD), Speech-to-Text (STT), Text-to-Speech (TTS), and integration with a Large Language Model (LLM). The assistant can be started in a room and interact with participants. 2. Frontend Options • Next.js Voice Agent Frontend: A flexible frontend for real-time voice interaction, video streaming, screen sharing, and more. • Mobile Applications: Starter templates are available for Android (Kotlin), iOS (Swift), Flutter, and React Native. 3. JavaScript/Node.js Implementation A Node.js implementation is also available, featuring similar capabilities as the Python version, including VAD, STT, TTS, and LLM integration. All implementations use the

VoicePipelineAgent

, which manages the core functionalities and can be customized for specific meeting call requirements. If you need further details or assistance with any of these options, let me know! Sources: templates.yaml | Recording agent sessions | LiveKit Docs | README.md | examples/src/pipeline_voice_agent.ts | VoicePipelineAgent | LiveKit Agents