Voice Processing

Voice Intelligent Technology (VIT) is our free, ready-to-use library that provides a far-field audio front end supporting different microphone topologies, a wake word engine and a voice command engine, with an online tool to generate customer-defined wake word and command models.

VoiceSpot is a low latency, ultra-low power wake word and command engine that performs well in noisy and high reverberation settings, using just a single microphone. It features an ultracompact memory footprint and low computational complexity.

VoiceSeeker is a multi-microphone audio front-end signal processing solution for low-power, always-on devices. It provides multi-mic beamforming, noise suppression, and multi-channel acoustic echo cancellation, enabling high performance far-field speech pickup. VoiceSeeker can be tightly integrated with VoiceSpot to provide excellent key word detection and payload capture in nearly any setting

Conversa is our advanced full-duplex voice processing software that leverages traditional and machine learning techniques to provide excellent speech quality in all noise types. It includes a real-time PC-based tuning tool that provides complete control of the audio signal path.