Use deep learning to reduce environmental noise and isolate voice.
- 500K~1MMemory Usage
- 32msLatency
- 3+Distortion Types: noise, reverb, band-limited and more
- 300M~1G MAC/sCompute
Our Edge
Strong speech restoration capability
Supports non-linear distortion and super-resolution in addition to denoising
Low-latency processing with configurable latency, model size, and denoise depth
Seamless integration with AudioWeaver and MATLAB
Industry-leading performance-to-footprint ratio