Direct Text to Speech Translation System using Acoustic Units
This work addresses the problem of translating text to speech without needing target language transcriptions, which is incremental as it builds on existing acoustic unit methods.
The paper tackles direct text-to-speech translation by using discrete acoustic units, achieving competitive performance on the CVSS corpus across most language pairs and showing significant improvement when initializing with a multilingual pre-trained model.
This paper proposes a direct text to speech translation system using discrete acoustic units. This framework employs text in different source languages as input to generate speech in the target language without the need for text transcriptions in this language. Motivated by the success of acoustic units in previous works for direct speech to speech translation systems, we use the same pipeline to extract the acoustic units using a speech encoder combined with a clustering algorithm. Once units are obtained, an encoder-decoder architecture is trained to predict them. Then a vocoder generates speech from units. Our approach for direct text to speech translation was tested on the new CVSS corpus with two different text mBART models employed as initialisation. The systems presented report competitive performance for most of the language pairs evaluated. Besides, results show a remarkable improvement when initialising our proposed architecture with a model pre-trained with more languages.