[FFmpeg-user] How to make ffmpeg keep silence in generated audio files when RTP input is paused

[Iñaki Baz Castillo]

On Wed, Sep 20, 2017 at 11:03 AM, Iñaki Baz Castillo <ibc at aliax.net> wrote:
> On Wed, Sep 20, 2017 at 10:30 AM, Iñaki Baz Castillo <ibc at aliax.net> wrote:
>> Thanks. May I know what AV_OPT_TYPE_DURATION is? seconds? ms?
>
> OK, it seems that the new -timeout N options is about seconds. I've
> seen that in other timeouts options -1 can be used to disable timeout.
> Not here. Any reason?

Well, I'm testing it. By setting -timeout 10000 (or bigger) it avoids
exiting the ffmpeg process when RTP is paused. However, If I run
ffmpeg when no RTP is being received, and the first RTP packet arrives
after 20 seconds, the generated output.wav does not start with 20
seconds of silence but, instead, with the first RTP packet.

This is not valid for me since in my scenario I must be able to launch
a ffmpeg recorder that generates silence in the output file even if no
audio/RTP is received (otherwise individual recordings of different
participants into a multiconference would not have the same start
time).




-- 
Iñaki Baz Castillo
<ibc at aliax.net>