How to output sentence-level text with timestamps?
#21
by aimagee - opened
Thanks for your great work. I'm using nyrahealth/CrisperWhisper model to transcribe audio to text with timestamps, but the model outputs word-level text with timestamps. How can I convert them to sentence-level timestamps?
Given that word-level is more granular than sentence-level, can't you just combine words until you see a pause or punctuation? eg [I, like, dogs, ] = "I like dogs." Then just take the start of the first word and the end of the last word for timestamps.