Use Platform-independent audio output

[Holzhaus]

AbstractSpeaker's play() method currently uses aplay for wav file playback. To make Jasper more platform independent, Jasper should use a more generic way of playing wav files.

Currently we can't get PyAudio to work, because this code may result in choppy playback in some configurations:

f = wave.open(filename, 'rb')
p = pyaudio.PyAudio()
stream = p.open(format=p.get_format_from_width(f.getsampwidth()),
                channels=f.getnchannels(),
                rate=f.getframerate(),
                output=True)
data = f.readframes(chunksize)
while data:
    stream.write(data)
    data = f.readframes(chunksize)
stream.stop_stream()
stream.close()
p.terminate()

Therefore, we switched back to aplay for now. Still, this is something we need to fix.

[Holzhaus]

The Evil line is here. I hope we can at least remove the -D hw:1,0 at some point, so that we no longer need to break our existing audio setups to use jasper.
By using hw:1,0, we also prevent user setups with different sound cards that do not support the sample rates we use in jasper by hardware. ALSA can be configured to do rate conversion automatically via dmix (or even via PulseAudio with device pulse). We break this possibility by hardcoding the hardware device.

[Holzhaus]

This will hopefully be fixed by #247.

[Holzhaus]

#247 was merged into the v2.0 dev branch, closing this.