.srt file as an input

[Madjid513]

Hey, I've been trying to implement a logic so that the model can handle .srt file by looping through each sentence from the .srt file then use another model to stretch it then combine it but it's just a mess and many errors keep occuring.

I'd like to know if there an easy way to use an .srt file as an input.

[vatsalaggarwal]

use another model to stretch it then combine it
not sure what you mean by this?

Are you trying to synthesise all the text from a given srt file (ignoring the time stamp information) as a long-form audio?

[Madjid513]

No i don't want to ingore time stamp.

1
00:00:00,140 --> 00:00:02,720
Voici mon conseil pour vous aujourd'hui : détournez-vous des 97 %

For example i want it to speak this sentence starting from 00:00:00,140 to 00:00:02,720 but since it wasn't possible i let the model speak it ignoring the time stamp then using another model that will speed up or slow down the speech but i wasn't able to make it work.