
The proposed USAT framework unifies both zero-shot and few-shot speaker adaptation strategies, outperforming state-of-the-art methods in synthesizing natural and speaker-similar speech for both native and non-native English speakers.


coremsg

universal-speaker-adaptive-text-to-speech-approach-outperforms-state-of-the-art-methods-across-native-and-non-native-english-speakers


Universal Speaker-Adaptive Text-to-Speech Approach Outperforms State-of-the-Art Methods Across Native and Non-Native English Speakers
