Using Phonemes in cascaded S2S translation pipeline
This work addresses translation efficiency and accessibility for low-resource language scenarios, though it is incremental as it adapts existing methods with a new representation.
The paper tackled the problem of speech-to-speech translation by using phonemes instead of text in a pipeline, finding that the phonemic approach achieved comparable BLEU scores while reducing resource needs and improving suitability for low-resource languages.
This paper explores the idea of using phonemes as a textual representation within a conventional multilingual simultaneous speech-to-speech translation pipeline, as opposed to the traditional reliance on text-based language representations. To investigate this, we trained an open-source sequence-to-sequence model on the WMT17 dataset in two formats: one using standard textual representation and the other employing phonemic representation. The performance of both approaches was assessed using the BLEU metric. Our findings shows that the phonemic approach provides comparable quality but offers several advantages, including lower resource requirements or better suitability for low-resource languages.