Enhancing expressivity transfer in textless speech-to-speech translation
This addresses the challenge of conveying emotions and nuances across languages for users of speech translation systems, though it is incremental as it builds on existing self-supervised learning techniques.
The study tackled the problem of poor expressivity transfer in textless speech-to-speech translation by introducing a method using multilingual emotion embeddings at the discrete speech unit level, resulting in superior expressivity transfer compared to state-of-the-art systems in French-to-English translation tasks.
Textless speech-to-speech translation systems are rapidly advancing, thanks to the integration of self-supervised learning techniques. However, existing state-of-the-art systems fall short when it comes to capturing and transferring expressivity accurately across different languages. Expressivity plays a vital role in conveying emotions, nuances, and cultural subtleties, thereby enhancing communication across diverse languages. To address this issue this study presents a novel method that operates at the discrete speech unit level and leverages multilingual emotion embeddings to capture language-agnostic information. Specifically, we demonstrate how these embeddings can be used to effectively predict the pitch and duration of speech units in the target language. Through objective and subjective experiments conducted on a French-to-English translation task, our findings highlight the superior expressivity transfer achieved by our approach compared to current state-of-the-art systems.