VoiceMOS Challenge 2026 tests speech quality and emotion ratings
The VoiceMOS Challenge 2026: Evaluating Speech Enhancement, Emotional TTS and Accented TTS Systems
Sound
Summary
This challenge evaluates how well computer programs can predict people’s opinions about different types of speech recordings. The authors organized three testing areas: judging the quality of improved speech, checking how natural and emotional computer-generated speech sounds, and comparing speaker accents using special speech codecs. Many teams participated and beat the basic benchmarks, showing progress in these areas. The authors share the challenge results, top methods, and ideas for next steps.
What this means in practice
- •For speech technology developers: Evaluate and improve automatic metrics that predict human opinions on enhanced and synthesized speech quality and emotion.
- •For call center software teams: Assess emotional and accented synthetic voices used in customer interactions for better naturalness and speaker matching.
Authors
Wen-Chin Huang, Wei Wang, Marvin Sach, Xiaoxue Gao, Nicholas Sanders, Erica Cooper, Toda Tomoki
Abstract
We present the results of the VoiceMOS Challenge 2026, the fifth edition of a scientific challenge on automatic prediction of subjective speech assessments. After expanding the scope to music and general audio in 2025, we refocused the evaluation target on speech and organized three tracks: prediction of absolute and comparative category ratings for enhanced speech, prediction of naturalness and emotion-related tasks for emotional text-to-speech systems, and prediction of speaker and accent similarity for codec-based speech synthesis systems. The challenge attracted a total of 18 teams worldwide, with most teams successfully surpassing the provided baselines. We summarize the challenge results, representative top-performing systems, participant feedback, and directions for future editions.