izwi-ai/izwi
On-device AI engine for transcription, TTS, and voice workflows.
Runs entirely on-device with no cloud dependencies, exposing OpenAI-compatible API endpoints for seamless integration. Supports advanced audio workflows including speaker diarization, voice cloning, voice design from text descriptions, and forced word-level alignment—all powered by a unified model ecosystem (Qwen3, Parakeet, Whisper, Gemma). Long-form ASR automatically chunks and stitches overlapping transcripts with configurable parameters, eliminating model window limitations.
181 stars.
Stars
181
Forks
18
Language
Rust
License
—
Category
Last pushed
Mar 18, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/izwi-ai/izwi"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
TrevorS/voxtral-mini-realtime-rs
Streaming speech recognition running natively and in the browser. A pure Rust implementation of...
mrtozner/vox
Local voice AI framework for Rust. Whisper + LLM + TTS with no cloud dependencies.
darkautism/sensevoice-rs
A Rust-based, SenseVoiceSmall
thewh1teagle/vad-rs
Speech detection using silero vad in Rust
0xPD33/sonori
Sonori is a fully local STT app for Linux (Wayland).