Request: Add argument to specify minimum buffer (# milliseconds) before fast-forwarding starts or ends about unsilence HOT 5 CLOSED

Hey, if I understood you right, unsilenced already has an argument for that:

parser.add_argument("-st", "--stretch-time", type=float, default=0.25, help="Time (seconds) that should be added to audible intervals and removed from silent " "intervals")

Here, before an audible interval starts, 250ms of the previous considered silent interval is made audible, and 250ms of the interval after the audible interval is considered audible. You can change the the time added to the audible intervals by changing the -st argument, it's value is given in seconds.

from unsilence.

Thanks for the help! That sounds like what I was looking for with this feature request; I'll test it for some files later today and see if it fixes, then I should be able to comment close.

Stretch-time is a single buffer for both the transition from silent to audible and back right? I could see reasons why you might want a separate argument for both.

from unsilence.

Stretch-time basically adds a buffer of x seconds before and after any audible inverval (maybe I should rename the argument, buffer is much more meaningfull than stretch-time). What's the reason you want to seperate the timings for each?

from unsilence.

My request for a separate buffer might be solved by the -stt argument you mentioned in my other feature request thread. I'll do some testing today and see if that covers my use case; it's possible the existing -st buffer will work when combined with -stt.

from unsilence.

Yep, the -stt argument solved my use case. Thank you for this! This tool is saving me hundreds upon hundreds of hours of what used to be dreadful manual editing work.

from unsilence.