audio codec

Satin is a neural speech codec developed to efficiently compress and transmit speech audio using deep learning techniques. It aims to provide high-quality voice communication at low bitrates by leveraging neural network models for encoding and decoding speech signals.