DualCodec is a low-frame-rate (12.5Hz or 25Hz), semantically-enhanced (with SSL feature) Neural Audio Codec designed to extract discrete tokens for efficient speech ...

Abstract: Automatic detection of synthetic speech is becoming increasingly important as current synthesis methods are both near indistinguishable from human speech and widely accessible to the public.

Abstract: End-to-end image and video codecs are becoming increasingly competitive, compared to traditional compression techniques that have been developed through decades of manual engineering efforts ...

Lossless audio is the first step toward audio nirvana. But what is it, does it really make a difference, and how can you get it? Here’s what to know. There’s a difference, of course, between “putting ...

This is the new updated repository. More information will be posted in the future. Currently nothing is functional. Feel free to check out in the future when stuff develop. For any contributions/help, ...