Exploring Whisper by OpenAI, wanted to work on one speech-to-text idea for a long. This model appears lucrative to me. 

Any comments or suggestions are welcome. Thanks