Saved in:
Bibliographic Details
Main Authors: Zhou, Haoguang, Wang, Siyi, Wu, Jingyao, Bailey, James, Dang, Ting
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.21224
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914413614727168
author Zhou, Haoguang
Wang, Siyi
Wu, Jingyao
Bailey, James
Dang, Ting
author_facet Zhou, Haoguang
Wang, Siyi
Wu, Jingyao
Bailey, James
Dang, Ting
contents Modern speech systems increasingly use discretized self-supervised speech representations for compression and integration with token-based models, yet their impact on emotional information remains unclear. We study how residual vector quantization (RVQ) reshapes emotional information in discrete speech representations from both representation- and task-level perspectives. Our analysis shows that aggressive compression disproportionately degrades emotion, with uneven loss across emotion classes and model architectures. To address this, we introduce emotion-aware quantization using emotion-specific and emotion-biased codebooks, improving the preservation of both hard and soft emotion perception. We further propose Emo-Q, a lightweight routed quantization method that selects emotion-specialized codebooks, improving emotion recognition performance at lower bitrates. These results highlight the importance of emotion-aware discretization for robust affective speech processing.
format Preprint
id arxiv_https___arxiv_org_abs_2603_21224
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Emotion-Aware Quantization for Discrete Speech Representations: An Analysis of Emotion Preservation
Zhou, Haoguang
Wang, Siyi
Wu, Jingyao
Bailey, James
Dang, Ting
Sound
Modern speech systems increasingly use discretized self-supervised speech representations for compression and integration with token-based models, yet their impact on emotional information remains unclear. We study how residual vector quantization (RVQ) reshapes emotional information in discrete speech representations from both representation- and task-level perspectives. Our analysis shows that aggressive compression disproportionately degrades emotion, with uneven loss across emotion classes and model architectures. To address this, we introduce emotion-aware quantization using emotion-specific and emotion-biased codebooks, improving the preservation of both hard and soft emotion perception. We further propose Emo-Q, a lightweight routed quantization method that selects emotion-specialized codebooks, improving emotion recognition performance at lower bitrates. These results highlight the importance of emotion-aware discretization for robust affective speech processing.
title Emotion-Aware Quantization for Discrete Speech Representations: An Analysis of Emotion Preservation
topic Sound
url https://arxiv.org/abs/2603.21224