-
Notifications
You must be signed in to change notification settings - Fork 1.6k
wav2vec German model added #1557
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Huge thanks for the PR!! Always great to have new languages in SpeechBrain 😁
Could you please update the README and add German too? (maybe you can add there the time taken per epoch)
If you agree, I will create a new model card on hugging face with this model when everything is ready. Then, you will have to make a pull request to add your model etc... 🤗
About what you did for the colouration in core.py, we will need more discussion on it. Could you please undo the core change with colour and create a separate pull request on that?
Your code failed on some tests (pre-commit and test), please run the pre-commit and tests as described here 🙂
Make sure to add your recipe in tests/recipes.csv to fix the test issue
Last but not least, could you please make sure that your .yaml is in the same order as the other .yaml (e.g French, Italian...)? For instance, you defined on line 104 the wav2vec2_hub while in the other yaml it is done on line 17. Could you please make sure that everything is similar?
|
c0f7b66 to
80fb945
Compare
|
@Adel-Moumen Changes are ready to be reviewed. |
Adel-Moumen
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hello,
Thanks for the changes!
There are some minor edits that I would like you to make, please. I will also create your HuggingFace card for this model asap, you will then need to update the link for the model as well.
Many thanks for your excellent work!
4d9c7b2 to
51f60fc
Compare
|
Apologies @Moumeneb1 , I had to close this PR, and re-open it because it got messed up when I pulled the latest changes today. |
Hello, you pinged the wrong reviewer Ahah. There is one last thing that needs to be done is the upload of your model on the HuggingFace hub. For now, we will need to wait a little bit more. Unfortunately, the process takes a little bit of time. |
|
Sure thing. |
|
@Adel-Moumen I was wondering if we have any updates further? ~Thanks |
|
Hello, It will be ready next week! :) |
|
Hello @sangeet2020, Could you please resolve the conflict with the I did not finish the upload on the HF model hub. I struggled to host your model on the hub due to the tokenizer used. If you agree, I can add you to the HF SpeechBrain organisation and then you will have to create a PR on the model card to add everything necessary. Let me know what you think! Thanks. :-) |
|
@sangeet2020, could you please solve the conflit? Also, it could be great if @anautsch could give a final look before merging it |
|
Re-opening the PR, fixing conflicts with |
|
Now, I can host the new German CommonVoice model on HF. You can add me to the And apologies for the delay, bit busy with a new recipe for Microsoft DNS. Should be on SpeechBrain soon. :) |
|
Hello, I added you to the HF organization. You can create a private repository on the organization, and start working on it! Let me know if you need more help. (you can reach me directly on slack) Please fix the recipes :) |
Thank You |
Adel-Moumen
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hello,
Thanks a lot! I tried your model and it's working perfectly.
Could you please update the readme, and recipe.csv with the model link, then we will merge the PR! :)
Adel-Moumen
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
everything looks great! thanks!
|
Thank you very much!! |
|
Thanks for contributing. |
|
@Adel-Moumen please check the model in the SpeechBrain's google drive link. You uploaded the model dir named Thank You |
|
Hi, please accept me on the gdrive thanks. |
Editing access granted. |
|
Thanks. I will edit the folder tomorrow! |
|
Hello @sangeet2020, Is it better like that? :-) |
|
Yes, this looks alright now :). Thank you very much. |
I initiate a pull request with the following additions, modifications and fixings:
huggingface/speechbrain.speechbrain/processing/features.py, a bug has been fixed as proposed in Device error when settingdeltas=Trueinspeechbrain.lobes.features.Fbank. #1489Thank You