-
Thank you for organizing this repo. The script is great.
However, I found that the `commonvoice_19.0_en_train_track2.json.gz` is missing.
As a result, dataprep for track2 won't run.
Can you fix thi…
-
Hi,
Kabyle model needed. We have a dataset of more than 698 hours recorded on both Mozilla CommonVoice and Tatoeba for Kabyle language `kab`.
Regards,
-
https://commonvoice.mozilla.org/ja
-
I'm trying to learn a Conformer model using data and a custom word dictionary.
My learning environment uses two GPUs, and the dataset uses commonvoice data.
Looking at this log, can i see that lea…
-
Hi,
What should I do for commonvoice?
-
## Support for Vietnamese Language
- Hi, I found 100 hours of voice data from VinBigdata - https://institute.vinbigdata.org/events/vinbigdata-chia-se-100-gio-du-lieu-tieng-noi-cho-cong-dong/
- H…
-
Research and verify the possibility of training a DeepSeach model for Brazilian Portuguese.
Related docs:
- https://deepspeech.readthedocs.io/en/r0.9/TRAINING.html
- https://commonvoice.mozilla.o…
-
### Bug report
PHPStan complains this function returns an array (not necessarily a list):
```
/**
* @return list
*/
function get_cv_files(): array
{
$pdo = new PDO("");
$stmt = $p…
-
how to handle big downloads?
do the sentences have a unique persistent ID?
any API for lookup using the ID?
emailed mozilla 2 days ago to ask
-
**Describe the bug**
The header text on the About page is misaligned on small screens (devices with width below 991px). It lacks proper padding, resulting in the text being cramped toward the left si…