Hirofumi Inaguma

https://hirofumi0810.github.io/ [email protected]

FAIR NYC Research scientist at FAIR

Results 7 issues of


                                            Hirofumi Inaguma

Multilingual End-to-End Speech Translation

6

Recipe

ESPnet1

Attention-based models do not work

3

I'm refactoring attention-based models, so they do not work now! Please wait for a few days.

Speed up by using LSTMBlockFused.

Reformat speech translation modules

# Before submitting - [ ] Was this discussed/approved via a Github issue? (no need for typos, doc improvements) - [x] Did you read the [contributor guideline](https://github.com/pytorch/fairseq/blob/main/CONTRIBUTING.md)? - [ ]...

CLA Signed

UnitY implementation

# Before submitting - [ ] Was this discussed/approved via a Github issue? (no need for typos, doc improvements) - [x] Did you read the [contributor guideline](https://github.com/pytorch/fairseq/blob/main/CONTRIBUTING.md)? - [ ]...

CLA Signed

Unify build_embedding

# Before submitting - [ ] Was this discussed/approved via a Github issue? (no need for typos, doc improvements) - [x] Did you read the [contributor guideline](https://github.com/pytorch/fairseq/blob/main/CONTRIBUTING.md)? - [ ]...

CLA Signed

Fix Must-C data preprocessing

# Before submitting - [ ] Was this discussed/approved via a Github issue? (no need for typos, doc improvements) - [x] Did you read the [contributor guideline](https://github.com/pytorch/fairseq/blob/main/CONTRIBUTING.md)? - [ ]...

CLA Signed