Is there a function for getting the timestamp for the middle when the word was spoken? #50
Unanswered
santiagodiaz1993
asked this question in
Q&A
Replies: 1 comment
-
The current implementation only gives the approximate end-timestamp of each word. The rest of the logic just attempts to make sense of that to display. The involves using the end-timestamps of the previous word as the start-timestamps of the current word. The "gaps" used to make the timestamps more granular are just parts of the audio that are completely silent but it doesn't mean the non-gaps always have speech in it. I think VAD would be more useful if you are just looking for the gaps between the speech. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi there, I would like to find the time of the silence gaps between words that are said in a script (A descent approximation should be enough). I used def results_to_word_srt but this returns approximations of when the words were said but does not take into account the silence in between words (the end timestamp for each word matches with the beginning timestamp of the next).
Is there a way to get the timestamp of the middle of when the word was said?. I think I could use this to figure out the silence gaps between words. The image in the description seems to show this.
Or is there another function that I can use to find the timestamps of the silence gaps?
Beta Was this translation helpful? Give feedback.
All reactions