Rethinking Mean Opinion Scores in Speech Quality Assessment: Aggregation through Quantized Distribution Fitting

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Kondo, Yuto, Kameoka, Hirokazu, Tanaka, Kou, Kaneko, Takuhiro
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916806302629888
author Kondo, Yuto
Kameoka, Hirokazu
Tanaka, Kou
Kaneko, Takuhiro
author_facet Kondo, Yuto
Kameoka, Hirokazu
Tanaka, Kou
Kaneko, Takuhiro
contents Speech quality assessment (SQA) aims to evaluate the quality of speech samples without relying on time-consuming listener questionnaires. Recent efforts have focused on training neural-based SQA models to predict the mean opinion score (MOS) of speech samples produced by text-to-speech or voice conversion systems. This paper targets the enhancement of MOS prediction models' performance. We propose a novel score aggregation method to address the limitations of conventional annotations for MOS, which typically involve ratings on a scale from 1 to 5. Our method is based on the hypothesis that annotators internally consider continuous scores and then choose the nearest discrete rating. By modeling this process, we approximate the generative distribution of ratings by quantizing the latent continuous distribution. We then use the peak of this latent distribution, estimated through the loss between the quantized distribution and annotated ratings, as a new representative value instead of MOS. Experimental results demonstrate that substituting MOSNet's predicted target with this proposed value improves prediction performance.
format Preprint
id arxiv_https___arxiv_org_abs_2506_18307
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Rethinking Mean Opinion Scores in Speech Quality Assessment: Aggregation through Quantized Distribution Fitting
Kondo, Yuto
Kameoka, Hirokazu
Tanaka, Kou
Kaneko, Takuhiro
Sound
Audio and Speech Processing
Speech quality assessment (SQA) aims to evaluate the quality of speech samples without relying on time-consuming listener questionnaires. Recent efforts have focused on training neural-based SQA models to predict the mean opinion score (MOS) of speech samples produced by text-to-speech or voice conversion systems. This paper targets the enhancement of MOS prediction models' performance. We propose a novel score aggregation method to address the limitations of conventional annotations for MOS, which typically involve ratings on a scale from 1 to 5. Our method is based on the hypothesis that annotators internally consider continuous scores and then choose the nearest discrete rating. By modeling this process, we approximate the generative distribution of ratings by quantizing the latent continuous distribution. We then use the peak of this latent distribution, estimated through the loss between the quantized distribution and annotated ratings, as a new representative value instead of MOS. Experimental results demonstrate that substituting MOSNet's predicted target with this proposed value improves prediction performance.
title Rethinking Mean Opinion Scores in Speech Quality Assessment: Aggregation through Quantized Distribution Fitting
topic Sound
Audio and Speech Processing
url https://arxiv.org/abs/2506.18307