yukaorigenji / dis_translatotron Goto Github PK

Can you upload that part or explain it. How was the performance compared to original translatotron

Hello, your project has not been introduced in detail, and I have never used tensorflow, so I am not sure about the meaning of your code.
Google published a paper titled Direct speech-to-speech translation with a sequence-to-sequence model in 2019, which can be obtained from the following hyperlink: https://arxiv.org/pdf/1904.06037v1. pdf. I am curious if your project is an attempt to accurately reproduce the results of this paper? If so, what are your results?
Or, to put it simply, I hope you can explain what your code achieves?
Looking forward to your reply. Your clear answer will be of great help to me.