Skip to content
This repository

Output PCM audio data to the speakers

branch: master

index: make _format() bind to the speaker instance

This way, we don't have to `delete` the callback function in
_unpipe() and can potentially reuse it...
latest commit f41ae829a6
Nathan Rajlich authored April 17, 2014
Octocat-spinner-32 deps mpg123: fix bad source file name November 03, 2012
Octocat-spinner-32 examples examples: fix "sine" emitting "end" event April 11, 2014
Octocat-spinner-32 src src: use `NanDisposePersistent()` April 17, 2014
Octocat-spinner-32 test throw an Error if non-native endianness is specified February 10, 2013
Octocat-spinner-32 .gitignore initial commit October 23, 2012
Octocat-spinner-32 .travis.yml travis: test node v0.11 April 11, 2014
Octocat-spinner-32 v0.1.0 April 17, 2014
Octocat-spinner-32 README++ April 13, 2014
Octocat-spinner-32 binding.gyp use rvagg/nan April 11, 2014
Octocat-spinner-32 index.js index: make _format() bind to the speaker instance April 17, 2014
Octocat-spinner-32 package.json v0.1.0 April 17, 2014


Output PCM audio data to the speakers

Build Status

A Writable stream instance that accepts PCM audio data and outputs it to the speakers. The output is backed by mpg123's audio output modules, which in turn use any number of audio backends commonly found on Operating Systems these days.


Simply compile and install node-speaker using npm:

$ npm install speaker

On Debian/Ubuntu, the ALSA backend is selected by default, so be sure to have the alsa.h header file in place:

$ sudo apt-get install libasound2-dev


Here's an example of piping stdin to the speaker, which should be 2 channel, 16-bit audio at 44,100 samples per second (a.k.a CD quality audio).

var Speaker = require('speaker');

// Create the Speaker instance
var speaker = new Speaker({
  channels: 2,          // 2 channels
  bitDepth: 16,         // 16-bit samples
  sampleRate: 44100     // 44,100 Hz sample rate

// PCM data from stdin gets piped into the speaker


require('speaker') directly returns the Speaker constructor. It is the only interface exported by node-speaker.

new Speaker([ format ]) -> Speaker instance;

Creates a new Speaker instance, which is a writable stream that you can pipe PCM audio data to. The optional format object may contain any of the Writable base class options, as well as any of these PCM formatting options:

  • channels - The number of audio channels. PCM data must be interleaved. Defaults to 2.
  • bitDepth - The number of bits per sample. Defaults to 16 (16-bit).
  • sampleRate - The number of samples per second per channel. Defaults to 44100.
  • signed - Boolean specifying if the samples are signed or unsigned. Defaults to true when bit depth is 8-bit, false otherwise.
  • samplesPerFrame The number of samples to send to the audio backend at a time. You likely don't need to mess with this value. Defaults to 1024.

"open" event

Fired when the backend open() call has completed. This happens once the first write() call happens on the speaker instance.

"flush" event

Fired after the speaker instance has had end() called, and after the audio data has been flushed to the speakers.

"close" event

Fired after the "flush" event, after the backend close() call has completed. This speaker instance is essentially finished after this point.

Audio Backend Selection

node-speaker is backed by mpg123's "output modules", which in turn use one of many popular audio backends like ALSA, OSS, SDL, and lots more. The default backends for each operating system are described in the table below:

Operating System Audio Backend Description
Linux alsa Output audio using Advanced Linux Sound Architecture (ALSA).
Mac OS X coreaudio Output audio using Mac OS X's CoreAudio.
Windows win32 Audio output for Windows (winmm).
Solaris sun Audio output for Sun Audio.

To manually override the default backend, pass the --mpg123-backend switch to npm/node-gyp:

$ npm install speaker --mpg123-backend=openal
Something went wrong with that request. Please try again.