Music player Python module

This Python module provides a high-level core Music player interface where you are supposed to provide all the remaining high-level logic like the user interface, the playlist logic and the audio data.

Example

A very simple player with gapless playback:

import musicplayer, sys, os, fnmatch, random, pprint, Tkinter

class Song:
    def __init__(self, fn):
        self.url = fn
        self.f = open(fn)
    # `__eq__` is used for the peek stream management
    def __eq__(self, other):
        return self.url == other.url
    # this is used by the player as the data interface
    def readPacket(self, bufSize):
        return self.f.read(bufSize)
    def seekRaw(self, offset, whence):
        r = self.f.seek(offset, whence)
        return self.f.tell()

files = []
def getFiles(path):
    for f in sorted(os.listdir(path), key=lambda k: random.random()):
        f = os.path.join(path, f)
        if os.path.isdir(f): getFiles(f) # recurse
        if len(files) > 1000: break # break if we have enough
        if fnmatch.fnmatch(f, '*.mp3'): files.append(f)
getFiles(os.path.expanduser("~/Music"))
random.shuffle(files) # shuffle some more

i = 0

def songs():
    global i, files
    while True:
        yield Song(files[i])
        i += 1
        if i >= len(files): i = 0

def peekSongs(n):
    nexti = i + 1
    if nexti >= len(files): nexti = 0
    return map(Song, (files[nexti:] + files[:nexti])[:n])

# Create our Music Player.
player = musicplayer.createPlayer()
player.outSamplerate = 96000 # support high quality :)
player.queue = songs()
player.peekQueue = peekSongs

# Setup a simple GUI.
window = Tkinter.Tk()
window.title("Music Player")
songLabel = Tkinter.StringVar()

def onSongChange(**kwargs): songLabel.set(pprint.pformat(player.curSongMetadata))
def cmdPlayPause(*args): player.playing = not player.playing
def cmdNext(*args): player.nextSong()

Tkinter.Label(window, textvariable=songLabel).pack()
Tkinter.Button(window, text="Play/Pause", command=cmdPlayPause).pack()
Tkinter.Button(window, text="Next", command=cmdNext).pack()

player.onSongChange = onSongChange
player.playing = True # start playing
window.mainloop()

Description

It provides a player object which represents the player. It needs a generator player.queue which yields Song objects which provide a way to read file data and seek in the file. See the source code for further detailed reference.

It has the following functionality:

• open source (simplified BSD license, see License.txt)
• very simple interface
• support of most important sound formats (MP3, Flac, Ogg Vorbis, WMA, AAC / ALAC m4a, ...)
• Plays audio data via the player object. Uses FFmpeg for decoding and PortAudio for playing.
• Of course, the decoding and playback is done in seperate threads. You can read about that here.
• Supports any sample rate via player.outSamplerate. The preferred sound device is set via player.preferredSoundDevice. Get a list of all sound devices via getSoundDevices().
• Can modify the volume via player.volume and also song.gain (see source code for details).
• Prevents clipping via a smooth limiting functions which still leaves most sounds unaffected and keeps the dynamic range (see smoothClip).
• ReplayGain (for audio volume normalization) (see pyCalcReplayGain). This is as far as I know the only other implementation of ReplayGain despite the original from mp3gain (gain_analysis.c).
• AcoustId audio fingerprint (see pyCalcAcoustIdFingerprint). This one is also used by MusicBrainz. It uses the Chromaprint lib for implementation.
• Provides a simple way to access the song metadata.
• Provides a way to calculate a visual thumbnail for a song which shows the amplitude and the spectral centroid of the frequencies per time (see pyCalcBitmapThumbnail). Inspired by this project.
• Gapless playback

Usages

The main usage is probably in the MusicPlayer project - a full featured high-quality music player.

Installation

To get the source working, you need these requirements:

• boost >=1.55.0
• ffmpeg >= 2.0 (including libswresample)
• portaudio >=v19
• chromaprint

Debian/Ubuntu

apt-get install python-dev libsnappy-dev libtool yasm libchromaprint-dev portaudio19-dev libboost-dev

FFmpeg:

apt-get install libavformat-dev libavresample-dev

If your FFmpeg in Debian/Ubuntu is too old (lacks libswresample), do:

add-apt-repository ppa:jon-severinsson/ffmpeg
apt-get update
apt-get install libavformat-dev libswresample-dev

or install it from source.

MacOSX

brew install boost
brew install portaudio
brew install ffmpeg
brew install chromaprint

Other notes

Chromaprint depends on FFmpeg, so if you have a custom FFmpeg install, you might also want to install that manually. ./configure && make && sudo make install should work for FFmpeg and PortAudio. You might also want to use --enable-shared for FFmpeg. cmake . && sudo make install for Chromaprint.)

Building

Then call python setup.py build or ./compile.py to build the Python modules (it will build the Python module musicplayer.so).

The module is also registered on PyPI, so you can also install via:

pip install musicplayer

Similar projects

• Overview in Python Wiki: Audio modules and Music software.
• PyAudio. MIT License. PortAudio wrapper. Thus, pretty low-level and no decoding functionality. Last update from 2012.
• PyFFmpeg. LGPL. FFmpeg wrapper. Thus, prettylow-level and no sound output. You could probably glue PyFFmpeg and PyAudio together for something useful but I expect it to be quite unstable and too slow. Basically, tis glue is done in C++ in this module.
• GStreamer Python Bindings. GStreamer is powerful but still too limited as a cross-platform music player backend solution. Quite heavy. That was my intuition. Maybe it's wrong and it would have been a perfect solution. But I think, in contrast, this module does a lot of things in a more compact and automatic/simpler way and at the same time provides more music player centric features.
• Beets. In its core, it is a music library manager and manages the metadata. It can calculate ReplayGain and AcoustID fingerprint. Via BPD plugin, it becomes a MPD compatible daemon player, based on GStreamer.
• libgroove. Library to provide a music player, very similar in functionality like this project, but pure C, no Python bindings.

Probably dead projects:

• PyMedia. LGPL, GPL. FFmpeg-based encoding/decoding of audio+video, sound input/output via OSS/Waveout/Wavein. Unfornutaley not well tuned for usage in a high-quality music player. Last update from 2006.
• Audiere. LGPL. High-level audio API, supports many sound formats and sound output on Windows/Linux. Last update from 2006.