Picovoice/cheetah
On-device streaming speech-to-text engine powered by deep learning
What it solves
Cheetah is an on-device streaming speech-to-text engine that allows developers to transcribe audio into text in real-time while keeping all voice processing local to the device for privacy and efficiency.
How it works
It operates as a cross-platform SDK that processes audio frames in real-time. Developers integrate the engine into their applications using a provided AccessKey for authentication. The engine processes audio streams and provides partial transcripts as it listens, with a final transcript generated once an endpoint (the end of a speech segment) is detected.
Who it’s for
Developers building applications for desktop (Windows, macOS, Linux), mobile (iOS, Android), web (Chrome, Safari, Firefox, Edge), and embedded devices (Raspberry Pi) who need high-performance, private, on-device transcription.
Highlights
- Privacy-First: All voice processing runs locally on the device.
- Broad Platform Support: Compatible with a wide range of OSs, browsers, and hardware including Raspberry Pi.
- Multi-Language Support: Supports English, French, German, Italian, Portuguese, and Spanish.
- Resource Efficient: Designed to be compact and computationally efficient for on-device deployment.