Investigating Neural Audio Codecs for Speech Language Model-Based Speech Generation
Abstract
Neural audio codec tokens serve as the fundamental building blocks for speech language model (SLM)-based speech generation. However, there is no systematic understanding on how the codec system affects the speech generation performance of the SLM. In this work, we examine codec tokens within SLM framework for speech generation to provide insights for effective codec design. We retrain existing high-performing neural codec models on the same data set and loss functions to compare their performance in a uniform setting. We integrate codec tokens into two SLM systems: masked-based parallel speech generation system and an auto-regressive (AR) plus non-auto-regressive (NAR) model-based system. Our findings indicate that better speech reconstruction in codec systems does not guarantee improved speech generation in SLM. A high-quality codec decoder is crucial for natural speech production in SLM, while speech intelligibility depends more on quantization mechanism.
Cite
@article{arxiv.2409.04016,
title = {Investigating Neural Audio Codecs for Speech Language Model-Based Speech Generation},
author = {Jiaqi Li and Dongmei Wang and Xiaofei Wang and Yao Qian and Long Zhou and Shujie Liu and Midia Yousefi and Canrun Li and Chung-Hsien Tsai and Zhen Xiao and Yanqing Liu and Junkun Chen and Sheng Zhao and Jinyu Li and Zhizheng Wu and Michael Zeng},
journal= {arXiv preprint arXiv:2409.04016},
year = {2024}
}
Comments
Accepted by SLT-2024