SecoustiCodec: Cross-Modal Aligned Streaming Single-Codecbook Speech Codec

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Qiang, Chunyu, Wang, Haoyu, Gong, Cheng, Wang, Tianrui, Fu, Ruibo, Wang, Tao, Chen, Ruilong, Yi, Jiangyan, Wen, Zhengqi, Zhang, Chen, Wang, Longbiao, Dang, Jianwu, Tao, Jianhua
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911091799359488
author Qiang, Chunyu
Wang, Haoyu
Gong, Cheng
Wang, Tianrui
Fu, Ruibo
Wang, Tao
Chen, Ruilong
Yi, Jiangyan
Wen, Zhengqi
Zhang, Chen
Wang, Longbiao
Dang, Jianwu
Tao, Jianhua
author_facet Qiang, Chunyu
Wang, Haoyu
Gong, Cheng
Wang, Tianrui
Fu, Ruibo
Wang, Tao
Chen, Ruilong
Yi, Jiangyan
Wen, Zhengqi
Zhang, Chen
Wang, Longbiao
Dang, Jianwu
Tao, Jianhua
contents Speech codecs serve as a crucial bridge in unifying speech and text language models. Existing codec methods face several challenges in semantic encoding, such as residual paralinguistic information (e.g., timbre, emotion), insufficient semantic completeness, limited reconstruction capability, and lack of support for streaming. To address these challenges, we propose SecoustiCodec, a cross-modal aligned low-bitrate streaming speech codec that disentangles semantic and paralinguistic information in a single-codebook space. To ensure semantic completeness and reconstruction fidelity, paralinguistic encoding is introduced to bridge the information gap between semantic and acoustic encoding. A semantic-only efficient quantization method based on VAE (Variational Autoencoder) and FSQ (Finite Scalar Quantization) is proposed. This approach alleviates the long-tail distribution problem of tokens while maintaining high codebook utilization. A semantic disentanglement method based on contrastive learning is proposed, which aligns text and speech in a joint multimodal frame-level space, effectively removing paralinguistic information from semantic encoding. An acoustic-constrained multi-stage optimization strategy is proposed to ensure robust and stable convergence. Figure~\ref{fig:pesq_kbps_below_2kbps} shows SecoustiCodec achieves SOTA (state-of-the-art) reconstruction quality (PESQ) of 1.77/2.58 at 0.27/1 kbps. The code and model weights for SecoustiCodec will be open-sourced upon the completion of the peer-review process. We've open-sourced SecoustiCodec's demo, code, and model weights.
format Preprint
id arxiv_https___arxiv_org_abs_2508_02849
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SecoustiCodec: Cross-Modal Aligned Streaming Single-Codecbook Speech Codec
Qiang, Chunyu
Wang, Haoyu
Gong, Cheng
Wang, Tianrui
Fu, Ruibo
Wang, Tao
Chen, Ruilong
Yi, Jiangyan
Wen, Zhengqi
Zhang, Chen
Wang, Longbiao
Dang, Jianwu
Tao, Jianhua
Audio and Speech Processing
Artificial Intelligence
Computation and Language
Sound
Speech codecs serve as a crucial bridge in unifying speech and text language models. Existing codec methods face several challenges in semantic encoding, such as residual paralinguistic information (e.g., timbre, emotion), insufficient semantic completeness, limited reconstruction capability, and lack of support for streaming. To address these challenges, we propose SecoustiCodec, a cross-modal aligned low-bitrate streaming speech codec that disentangles semantic and paralinguistic information in a single-codebook space. To ensure semantic completeness and reconstruction fidelity, paralinguistic encoding is introduced to bridge the information gap between semantic and acoustic encoding. A semantic-only efficient quantization method based on VAE (Variational Autoencoder) and FSQ (Finite Scalar Quantization) is proposed. This approach alleviates the long-tail distribution problem of tokens while maintaining high codebook utilization. A semantic disentanglement method based on contrastive learning is proposed, which aligns text and speech in a joint multimodal frame-level space, effectively removing paralinguistic information from semantic encoding. An acoustic-constrained multi-stage optimization strategy is proposed to ensure robust and stable convergence. Figure~\ref{fig:pesq_kbps_below_2kbps} shows SecoustiCodec achieves SOTA (state-of-the-art) reconstruction quality (PESQ) of 1.77/2.58 at 0.27/1 kbps. The code and model weights for SecoustiCodec will be open-sourced upon the completion of the peer-review process. We've open-sourced SecoustiCodec's demo, code, and model weights.
title SecoustiCodec: Cross-Modal Aligned Streaming Single-Codecbook Speech Codec
topic Audio and Speech Processing
Artificial Intelligence
Computation and Language
Sound
url https://arxiv.org/abs/2508.02849