CleanCodec: Efficient and Robust Speech Tokenization via Perceptually Guided Encoding 事件

CleanCodec: Efficient and Robust Speech Tokenization via Perceptually Guided Encoding arXiv:2606.04418v1 Announce Type: cross Abstract: Neural audio codecs are a key component of speech processing pipelines, compressing audio into discrete tokens for downstream modeling. However, existing codecs struggle to balance reconstruction quality with token efficiency, often encoding perceptually irrelevant information such as background noise and recording artifacts at the expense of linguistically and