Translatotron is our experimental model for direct end-to-end speech-to-speech translation, which demonstrates the potential for improved translation efficiency, fewer errors, and better handling of proper nouns. Learn all about it below!https://goo.gle/2Hlf5sQ
The encoding of language in wave forms to obviate the speech-to-text phase for MT is intriguing. Are you also seeking to encode intonation to attempt to convey emotion in the target language?