Saved in:
Bibliographic Details
Main Authors: Welford, Alejandro Sosa, Pepino, Leonardo
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2507.01805
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909673375924224
author Welford, Alejandro Sosa
Pepino, Leonardo
author_facet Welford, Alejandro Sosa
Pepino, Leonardo
contents This work addresses the development of a database for the automatic assessment of text-to-speech (TTS) systems in Spanish, aiming to improve the accuracy of naturalness prediction models. The dataset consists of 4,326 audio samples from 52 different TTS systems and human voices and is, up to our knowledge, the first of its kind in Spanish. To label the audios, a subjective test was designed based on the ITU-T Rec. P.807 standard and completed by 92 participants. Furthermore, the utility of the collected dataset was validated by training automatic naturalness prediction systems. We explored two approaches: fine-tuning an existing model originally trained for English, and training small downstream networks on top of frozen self-supervised speech models. Our models achieve a mean absolute error of 0.8 on a five-point MOS scale. Further analysis demonstrates the quality and diversity of the developed dataset, and its potential to advance TTS research in Spanish.
format Preprint
id arxiv_https___arxiv_org_abs_2507_01805
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Dataset for Automatic Assessment of TTS Quality in Spanish
Welford, Alejandro Sosa
Pepino, Leonardo
Sound
Audio and Speech Processing
This work addresses the development of a database for the automatic assessment of text-to-speech (TTS) systems in Spanish, aiming to improve the accuracy of naturalness prediction models. The dataset consists of 4,326 audio samples from 52 different TTS systems and human voices and is, up to our knowledge, the first of its kind in Spanish. To label the audios, a subjective test was designed based on the ITU-T Rec. P.807 standard and completed by 92 participants. Furthermore, the utility of the collected dataset was validated by training automatic naturalness prediction systems. We explored two approaches: fine-tuning an existing model originally trained for English, and training small downstream networks on top of frozen self-supervised speech models. Our models achieve a mean absolute error of 0.8 on a five-point MOS scale. Further analysis demonstrates the quality and diversity of the developed dataset, and its potential to advance TTS research in Spanish.
title A Dataset for Automatic Assessment of TTS Quality in Spanish
topic Sound
Audio and Speech Processing
url https://arxiv.org/abs/2507.01805