Open-Speech-EkStep / ULCA-asr-dataset-corpus

Creative Commons Attribution 4.0 International
38 stars 16 forks source link

Vakyansh models are fine-tuned on much less/more labelled data (transcribed audios). So why there is a discrepancy? #2

Open ritwikmishra opened 2 years ago

ritwikmishra commented 2 years ago

In this repo the size of Hindi labelled data is mentioned as 4200 hours but this repository contains Hindi labelled data with duration of 2398 hours.

On the contrary, Tamil labelled data in this repository (1160 hours) is much more than Tamil data (250 hour) on which Vakyansh model for Tamil has been fine-tuned.

Similarly for other languages

So why there is discrepancy?