Saved in:
Bibliographic Details
Main Authors: Li, Jiaqi, Qian, Yao, Hu, Yuxuan, Zhang, Leying, Wang, Xiaofei, Lu, Heng, Thakker, Manthan, Li, Jinyu, Zhao, Sheng, Wu, Zhizheng
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.00981
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908851540852736
author Li, Jiaqi
Qian, Yao
Hu, Yuxuan
Zhang, Leying
Wang, Xiaofei
Lu, Heng
Thakker, Manthan
Li, Jinyu
Zhao, Sheng
Wu, Zhizheng
author_facet Li, Jiaqi
Qian, Yao
Hu, Yuxuan
Zhang, Leying
Wang, Xiaofei
Lu, Heng
Thakker, Manthan
Li, Jinyu
Zhao, Sheng
Wu, Zhizheng
contents Neural audio codecs are foundational to speech language models. It is expected to have a low frame rate and decoupled semantic and acoustic information. A lower frame rate codec can reduce the computational cost of speech language models by shortening the sequence length. Recent studies have developed 12.5Hz low-frame-rate audio codecs, but even lower frame rate codecs remain underexplored. We find that a major challenge for very low frame rate tokens is missing semantic information. This paper introduces FlexiCodec to address this limitation. FlexiCodec improves semantic preservation with a dynamic frame rate approach and introduces a novel architecture featuring an ASR feature-assisted dual stream encoding and Transformer bottlenecks. With dynamic frame rates, it uses less frames at information-sparse regions through adaptively merging semantically similar frames. A dynamic frame rate also allows FlexiCodec to support inference-time controllable frame rates between 3Hz and 12.5Hz. Experiments on 6.25Hz, 8.3Hz and 12.5Hz average frame rates confirm that FlexiCodec excels over baseline systems in semantic information preservation and delivers a high audio reconstruction quality. We also validate the effectiveness of FlexiCodec in language model-based TTS. Demos are available at: https://flexicodec.github.io. Code is available at: https://github.com/amphionteam/flexicodec.
format Preprint
id arxiv_https___arxiv_org_abs_2510_00981
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FlexiCodec: A Dynamic Neural Audio Codec for Low Frame Rates
Li, Jiaqi
Qian, Yao
Hu, Yuxuan
Zhang, Leying
Wang, Xiaofei
Lu, Heng
Thakker, Manthan
Li, Jinyu
Zhao, Sheng
Wu, Zhizheng
Sound
Neural audio codecs are foundational to speech language models. It is expected to have a low frame rate and decoupled semantic and acoustic information. A lower frame rate codec can reduce the computational cost of speech language models by shortening the sequence length. Recent studies have developed 12.5Hz low-frame-rate audio codecs, but even lower frame rate codecs remain underexplored. We find that a major challenge for very low frame rate tokens is missing semantic information. This paper introduces FlexiCodec to address this limitation. FlexiCodec improves semantic preservation with a dynamic frame rate approach and introduces a novel architecture featuring an ASR feature-assisted dual stream encoding and Transformer bottlenecks. With dynamic frame rates, it uses less frames at information-sparse regions through adaptively merging semantically similar frames. A dynamic frame rate also allows FlexiCodec to support inference-time controllable frame rates between 3Hz and 12.5Hz. Experiments on 6.25Hz, 8.3Hz and 12.5Hz average frame rates confirm that FlexiCodec excels over baseline systems in semantic information preservation and delivers a high audio reconstruction quality. We also validate the effectiveness of FlexiCodec in language model-based TTS. Demos are available at: https://flexicodec.github.io. Code is available at: https://github.com/amphionteam/flexicodec.
title FlexiCodec: A Dynamic Neural Audio Codec for Low Frame Rates
topic Sound
url https://arxiv.org/abs/2510.00981