Beyond Naturalness: Probing Automated Text-To-Speech Evaluators on Linguistically Grounded Dimensions
By Oluwanifemi Bamgbose · Paper · cs.SD
Automated Text-to-Speech (TTS) evaluation methods (Mean Opinion Score (MOS) predictors and Audio Large Language Models (Audio-LLM) judges) are expected to reflect human perception, yet it is unclear how well they capture the distinct aspects of speech that listeners actually perc