Factorized RVQ-GAN For Disentangled Speech Tokenization

Factorized RVQ-GAN For Disentangled Speech Tokenization

Stať ve sborníku v databázi WoS či Scopus

We propose Hierarchical Audio Codec (HAC), a unified neural speech codec that factorizes its bottleneck into three linguistic levels-acoustic, phonetic, and lexical-within a single model. HAC leverages two knowledge distillation objectives: one from a pre-trained speech encoder (HuBERT) for phoneme-level structure, and another from a text-based encoder (LaBSE) for lexical cues. Experiments on English and multilingual data show that HAC's factorized bottleneck yields disentangled token sets: one aligns with phonemes, while another captures word-level semantics. Quantitative evaluations confirm that HAC tokens preserve naturalness and provide interpretable linguistic information, outperforming single-level baselines in both disentanglement and reconstruction quality. These findings underscore HAC's potential as a unified discrete speech representation, bridging acoustic detail and lexical meaning for downstream speech generation and understanding tasks.

We propose Hierarchical Audio Codec (HAC), a unified neural speech codec that factorizes its bottleneck into three linguistic levels-acoustic, phonetic, and lexical-within a single model. HAC leverages two knowledge distillation objectives: one from a pre-trained speech encoder (HuBERT) for phoneme-level structure, and another from a text-based encoder (LaBSE) for lexical cues. Experiments on English and multilingual data show that HAC's factorized bottleneck yields disentangled token sets: one aligns with phonemes, while another captures word-level semantics. Quantitative evaluations confirm that HAC tokens preserve naturalness and provide interpretable linguistic information, outperforming single-level baselines in both disentanglement and reconstruction quality. These findings underscore HAC's potential as a unified discrete speech representation, bridging acoustic detail and lexical meaning for downstream speech generation and understanding tasks.

Audio Codec | GAN | RVQ | Speech Tokenization

Audio Codec | GAN | RVQ | Speech Tokenization

KHURANA, S.; KLEMENT, D.; LAURENT, A.; BOBOS, D.; NOVOSAD, J.; GAZDIK, P.; ZHANG, E.; HUANG, Z.; HUSSEIN, A.; MARXER, R.; MASUYAMA, Y.; AIHARA, R.; HORI, C.; GERMAIN, F.; WICHERN, G.; LE ROUX, J.

01.01.2025

International Speech Communication Association

Rotterdam, The Netherlands

Proceedings of the Annual Conference of the International Speech Communication Association Interspeech

Interspeech

Nizozemsko

3514

3518

5

https://www.isca-archive.org/interspeech_2025/khurana25_interspeech.pdf

khurana_interspeech_2025