FreeCodec: A disentangled neural speech codec with fewer tokens

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Zheng, Youqiang, Tu, Weiping, Kang, Yueteng, Chen, Jie, Zhang, Yike, Xiao, Li, Yang, Yuhong, Ma, Long
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911026882019328
author Zheng, Youqiang
Tu, Weiping
Kang, Yueteng
Chen, Jie
Zhang, Yike
Xiao, Li
Yang, Yuhong
Ma, Long
author_facet Zheng, Youqiang
Tu, Weiping
Kang, Yueteng
Chen, Jie
Zhang, Yike
Xiao, Li
Yang, Yuhong
Ma, Long
contents Neural speech codecs have gained great attention for their outstanding reconstruction with discrete token representations. It is a crucial component in generative tasks such as speech coding and large language models (LLM). However, most works based on residual vector quantization perform worse with fewer tokens due to low coding efficiency for modeling complex coupled information. In this paper, we propose a neural speech codec named FreeCodec which employs a more effective encoding framework by decomposing intrinsic properties of speech into different components: 1) a global vector is extracted as the timbre information, 2) a prosody encoder with a long stride level is used to model the prosody information, 3) the content information is from a content encoder. Using different training strategies, FreeCodec achieves state-of-the-art performance in reconstruction and disentanglement scenarios. Results from subjective and objective experiments demonstrate that our framework outperforms existing methods.
format Preprint
id arxiv_https___arxiv_org_abs_2412_01053
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle FreeCodec: A disentangled neural speech codec with fewer tokens
Zheng, Youqiang
Tu, Weiping
Kang, Yueteng
Chen, Jie
Zhang, Yike
Xiao, Li
Yang, Yuhong
Ma, Long
Sound
Audio and Speech Processing
Neural speech codecs have gained great attention for their outstanding reconstruction with discrete token representations. It is a crucial component in generative tasks such as speech coding and large language models (LLM). However, most works based on residual vector quantization perform worse with fewer tokens due to low coding efficiency for modeling complex coupled information. In this paper, we propose a neural speech codec named FreeCodec which employs a more effective encoding framework by decomposing intrinsic properties of speech into different components: 1) a global vector is extracted as the timbre information, 2) a prosody encoder with a long stride level is used to model the prosody information, 3) the content information is from a content encoder. Using different training strategies, FreeCodec achieves state-of-the-art performance in reconstruction and disentanglement scenarios. Results from subjective and objective experiments demonstrate that our framework outperforms existing methods.
title FreeCodec: A disentangled neural speech codec with fewer tokens
topic Sound
Audio and Speech Processing
url https://arxiv.org/abs/2412.01053