TensorFlowASR icon indicating copy to clipboard operation
TensorFlowASR copied to clipboard

Published 20 hours ago • verified publisher icon TensorSpeech

Training ASR models in multiple languages

Open monatis opened this issue 4 years ago • 9 comments

TensorFlowASR makes it quite easy to train and deploy almost SOTA ASR models, but it provides a pretrained model only in English. On the other hand, FAIR has recently published an open and free dataset in 8 languages (see the paper). It is in the public domain and of a large size, and has the same quality as LibriSpeech. So, my suggestion is form a volunteer working group to collaborate on training ASR models in multiple languages and share them publically.

Maintainers of the repo can pin the issue and label it with help-wanted for visibility if this idea makes sense.

monatis avatar Dec 30 '20 08:12 monatis

This is the great idea :laughing:

nglehuy avatar Dec 30 '20 13:12 nglehuy

Hi @usimarit! Thanks for such a great project and your support to boost visibility of this issue 😻 So I will start by writing a helper script to automatically download MLS dataset in a given language and prepare the transcription and alphabet files and PR to add it to the repo. Then I can train a Conformer model in German by using this script as a first step 🚀

monatis avatar Dec 30 '20 14:12 monatis

Hi @monatis, just for your information, we should train using subwords instead of characters for performance boost

nglehuy avatar Dec 31 '20 09:12 nglehuy

Hi @usimarit, yeah I know that training with subwords yields a better performance, but I'm automatically generating an alphabet file in #92 for those who want to use characters anyway.

monatis avatar Dec 31 '20 09:12 monatis

Hi everyone, if you guys want to share your pretrained models, just upload the .h5 or .pb files along with config.yml to any drive (google, dropbox, etc.), then write a section, add the link, add your contact to a subsection in that section, in the README.md in the example directory that belongs to each model like in image (the one I made): Screenshot from 2021-01-26 19-42-51 And finally open a pull request to merge to the repo :smile:

nglehuy avatar Jan 26 '21 12:01 nglehuy

@usimarit Grat. I've been quite busy for some time, but I'll be more active in this repo on following days and contribute pretrained models in other languages. Thanks

monatis avatar Jan 26 '21 13:01 monatis

@monatis, did you already train a Conformer model in German?

JStumpp avatar Feb 04 '21 22:02 JStumpp

Hi @JStumpp I started to train it and hope to release it next week.

monatis avatar Feb 05 '21 06:02 monatis

Do you know in which subset of Librispeech is the english pretrained model trained on?

christina284 avatar Dec 15 '22 15:12 christina284