Translatotron 2: High-quality direct speech-to-speech translation with voice preservation. (arXiv:2107.08661v5 [cs.CL] UPDATED)

We present Translatotron 2, a neural direct speech-to-speech translation
model that can be trained end-to-end. Translatotron 2 consists of a speech
encoder, a linguistic decoder, an acoustic synthesizer, and a single attention
module that connects them together. Experimental results on three datasets
consistently show that Translatotron 2 outperforms the original Translatotron
by a large margin on both translation quality (up to +15.5 BLEU) and speech
generation quality, and approaches the same of cascade systems. In addition, we
propose a simple …

arxiv quality speech translation voice