---
date: 2016-06-22
title: ALSA, MPD and Pulseaudio output capturing
---

To be able to visualize my laptop's sound output with a python script, I needed an extra audio input that made it possible to capture the system's audio output.
<!--more-->

To achieve that, I did the following on my Raspberry Pi:

  * install alsa-utils and pulseaudio
  * get the pulseaudio init script [here](https://www.raspberrypi.org/forums/viewtopic.php?p=619713#p619713)
  * add the users to the group "pulse-access", I got `ALSA lib pulse.c:243:(pulse_connect) PulseAudio: Unable to connect` without it even as root
  * [create an ALSA monitor source](https://wiki.archlinux.org/index.php/PulseAudio/Examples#ALSA_monitor_source)

It is also possible to redirect audio to a FIFO.

MPD:

```
audio_output {
  type: "fifo"
  name: "FIFO"
  path: "/tmp/mpd.fifo"
  format: "44100:16:1"
}
```

Pulseaudio:

```bash
pacmd load-module module-pipe-sink file=/tmp/pulse.fifo
pacmd list-sinks  # search your default sink
pacmd load-module module-combine-sink slaves=yoursink,fifo_output
pacmd set-default-sink combined
```

ALSA:

```
pcm.writeFile {
  type file
  slave.pcm null
  file "/tmp/alsa.fifo"
  format "raw"
}
```

See also the [snapcast readme](https://github.com/badaix/snapcast).