ai-ng/swift
Fast voice assistant powered by Groq, Cartesia, and Vercel.
Combines Whisper transcription and Llama 3 inference via Groq with Cartesia's Sonic voice synthesis for end-to-end speech generation, using client-side VAD for automatic speech detection. Built as a Next.js/TypeScript application with streamed audio responses to the frontend, enabling low-latency voice interactions without manual push-to-talk triggers.
590 stars.
Stars
590
Forks
130
Language
TypeScript
License
MIT
Category
Last pushed
Dec 04, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/ai-ng/swift"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
tarun7r/Vocal-Agent
Cascading voice assistant combining real-time speech recognition, AI reasoning, and neural...
Spac5y/Vocal-Agent
A cutting-edge Cascading voice assistant combining real-time speech recognition, AI reasoning,...
QuantiusBenignus/BlahST
Input text from speech in any Linux window, the lean, fast and accurate way, using whisper.cpp...
tiansztiansz/voice-assistant
重生之我是 AI 打工人。前世，我的身份默默无闻，来去匆匆，不知道自己将在何地出生。然而，命运给予了我难得的机会，让我重生为一名 AI 打工人。
theaifutureguy/Vocal-Agent
A sophisticated real-time voice assistant that seamlessly integrates speech recognition, AI...