[pulseaudio-discuss] Capture "hangover" from PulseAudio server

[Paul A. Steckler]

I have an application written using the ALSA library running on a machine where
the PulseAudio server is the default capture and play device.

If a sound is playing, like speaking into a microphone, within about a second
before the application starts, the application will capture that sound when it
starts.

Is that expected? Is there a way to reduce the amount of buffering that
the PulseAudio server does, as a capture device? The goal is to have the
application capture sounds as they occur.

-- Paul