Deepspeech for Indian Language

I have around 55-60 hrs of Indian Speech (Common Voice[indian] + Indic TTS + youtube(8-9 hrs of data augmented to ~30 hrs)). I’m trying to train deepspeech with the dataset; however, I’m getting very poor output…

Sample Ground Truth: ‘This is lovely’
Sample res:‘l’

I was initially using 0.3.0 checkpoint. I’m also trying with 0.4.1 checkpoint. I’ve had around 10 epcoh (from checkpoint) with train batch 24, dev & test bach 8 each.

The res is always a single alphabet.

Any help would be appreciated. If anyone has worked on Indian Language, probably s/he can kindly give direction to proceed.

Any help would be greatly appreciated. Thank you.

any updates on this?
i got similar results for marathi.
how did you fix it?

Hi would you please explain your situation in a bit more detail? I might be able to help you out.

Hi @alchemi5t, I am trying to build a Hindi ASR using Deepspeech. If you have built one in Marathi could you help me out with the process.