Picovoice/cheetah
On-device streaming speech-to-text engine powered by deep learning
Delivers low-latency streaming transcription with sub-50ms latency on edge devices, requiring only an AccessKey for license validation while processing audio entirely offline. Supports six languages natively and runs across 15+ platforms including embedded systems (Raspberry Pi), mobile (iOS/Android), web browsers, and desktop environments via unified SDKs. Optimized for real-time performance with minimal computational overhead, making it suitable for privacy-sensitive voice applications without cloud dependencies.
661 stars. Actively maintained with 18 commits in the last 30 days.
Stars
661
Forks
76
Language
Python
License
Apache-2.0
Category
Last pushed
Mar 18, 2026
Commits (30d)
18
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/Picovoice/cheetah"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Compare
Related tools
PaddlePaddle/PaddleSpeech
Easy-to-use Speech Toolkit including Self-Supervised Learning model, SOTA/Streaming ASR with...
k2-fsa/sherpa
Speech-to-text server framework with next-gen Kaldi
Picovoice/leopard
On-device speech-to-text engine powered by deep learning
zaigie/FunSpeech
开箱即用的本地私有化部署语音服务，快速搭建FunASR与CosyVoice2/3后端
manyeyes/ManySpeech
AI Speech Solutions for Tasks such as ASR, Vocal Extraction, Accompaniment Extraction, Audio...