Draft for reading files in chunks

[Abhinay1997]

TODO:

From Zack:

The tricky part is keeping track of timestamps as we iterate through, there can be drift
If we just do every 30s we’re going to be missing some text because the model only transcribes up to a good breaking point, and assumes we will seek to that point and run the next 30s
Yea we just seek to them, which is where you could insert your method, instead of seeking, you load the next chunk and start at 0 time again

[Abhinay1997]

Putting on hold in favour of #125