WebRTC AI Voice Chat

Overview

The goal of this project is to demo speech <-> langchain <-> audio workflow.

1. Speech to text is using OpenAI's open source Whisper mini model.
2. Chat model used for this demo is Microsoft's Phi3 model running locally using Ollama.
3. Text to Audio is using Suno's open source Bark small model.

For interesting projects and related resources, checkout the Awesome Projects Page.

Demo

Unmute the audio to hear responses

https://github.com/lalanikarim/webrtc-ai-voice-chat/assets/1296705/7aa05d6f-ff05-4c72-b2e8-6e4e1119a68c