Model training on demand

Hi everyone,

I rented a server and am training models on demand for the rest of the month. I’ve done Chuvash (cv) and Tatar (tt) so far: https://tepozcatl.omnilingo.cc/

Here is a list of the rest of the language codes of the languages I plan to train: et, ta, tr, ky, dv, mn, id, br, mt, el, th, rm-sursilv, ro, hu, ia, sl, lv, lg, or, sah, cnh, ga-IE, ja, lt, rm-vallader, ka, hsb, pa-IN, vi, fi, ab, as, hi, vot.

If you would like a trained model for your language that is in this list, find me on Matrix #Common Voice and let me know and I’ll bump it up the priority list. :slight_smile:

3 Likes

The following models are now available: pt, et , ta , tr , ky , dv , mn , br , mt , lg , sah , cnh , ga-IE , ka , hsb, fi

The following models could unfortunately not be trained due to lack of data: ab , as , hi , vot.

The following models are due to be trained: id , el, th , rm-sursilv , ro , hu , ia , sl , lv , or , ja , ka , pa-IN , vi.

Great initiative, thanks for your work! Mozilla should do this after every dataset release for every language with more than 300 hours or so.

Would you mind sharing the statistics a little? Like the WER for every model.

I think they should do it after every language with more than one hour! They take hardly any time to train (about 30 minutes on an old GPU). The stats are online here: https://tepozcatl.omnilingo.cc/manifest.html

An interesting thing about this graph is that you can see orthographic effects :slight_smile:

As regards WER, the systems are not really usable at the moment, but they could easily be fine-tuned for e.g. closed-vocabulary tasks (I provide the checkpoints and the alphabet). What is shocking is that for some the CER gets close to around 20, which means that potentially they could be useful in applications like indexing/audio search. There are probably a lot of useful things that can be done with less-than perfect ASR…

The following models are due to be trained: rm-sursilv , ia , ja, vi, rm-vallader.

1 Like

If you plot the CER against seconds per character in the alphabet, then the effect is more clear:

1 Like

This is awesome, I wanna try some models as soon as possible.

1 Like

You can download them right now from here. :slight_smile: The Portuguese is one of the better ones. Not ready for use maybe, but ready for further finetuning.

I have uploaded a new one for Portuguese that is even better:


I’m still working on it, but will update here.