The problem Speech Recognition and Speaker Diarization solves

Speech Recognition Service for different accents and languages is Limited and by searching on how scrape and train your model to scale Speech Recognition to almost all different languages in India.
In this Notebook we scrape data from video sources to train deepspeech on different languages in google colab

Challenges I ran into

data collection and scraping was a challenge for indian english but then we got a rich dataset from IIT madras with different indian English accents from all parts of India