cheetah and leopard
Cheetah and Leopard are ecosystem siblings—Cheetah is optimized for real-time streaming speech-to-text while Leopard handles non-streaming (batch) audio processing, both built on Picovoice's platform for different use-case requirements.
About cheetah
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.
About leopard
Picovoice/leopard
On-device speech-to-text engine powered by deep learning
Supports real-time transcription across 8 languages with a lightweight model optimized for low-latency inference on edge devices (achieving real-time factor <1). Built with native SDKs for Python, Java, C, iOS, Android, Node.js, Flutter, React Native, and Web, enabling deployment from embedded systems (Raspberry Pi) to browsers with authentication via AccessKey.
Scores updated daily from GitHub, PyPI, and npm data. How scores work