The most cost-effective, highest performance AI voice agent possible today

Telephony Server is a powerful bridge that connects telephony providers (Twilio, Vonage, Plivo, etc.) with real-time communication platforms (LiveKit, Jay.so, Pipecat, etc.). It enables seamless call routing, robust metrics collection, and observability features for enhanced telephony operations.

Open Source Voice AI Dashboard

Quickstart template to build real-time voice AI solutions using the LiveKit Agent framework with WebRTC, open-source and local or 3rd party services such as AWS, Deepgram, Groq or OpenAI. Designed for low-latency.

A Streamlit web app for speaker diarization and identification in audio files. Upload or record audio, transcribe conversations, and automatically segment and label speakers using reference samples. This app makes it easy to analyze multi-speaker audio, export transcripts, and identify "who spoke when" for meetings, interviews, and more.

Demo applications using the Deepgram APIs in Rust. 🦀

An AI-powered platform for preparing for mock interviews with real-time feedback and personalized questions.

Detecting & Mitigating Bias: Gender, Age & Site

Client project for Fontys University of Applied Sciences. Semester 4 Creative Technology

A real-world AI-powered mock interview platform using Next.js, OpenAI GPT-4, and Vapi for voice-based LLM interactions. 🚀🎙️

YAPPER | Open Source Voice AI Dashboard

Voice AI Universal API to connect multiple Realtime voice providers like Gemini Live or OpenAi Realtime

An unofficial Rust API client for Vapi - Voice AI for developers.