Pre-PyConUK presentation tweaks

Switched over-sampling off.
Recording at 8k, playback at 11k.

End-of-speech detection

• Adds end-of-speech detection for more responsive playback
• Playback now uses dac.write()
• Documentation & global clean-up
• Adds oversampling mechanism
• Default settings are now:
  • 8kHz sampling
  • 8-bits
  • 7-second speech buffer
  • Attenuated
  • Automatic end-of-speech detection
  • Oversampling at 16kHz

Parameter and doc adjustments

• The code used for the final (live) demo at the presentation
• Fixed threshold parameters (a work in progress)
• 8kHz collection, 12kHz replay
• Minor documentation tweaks

Initial release

Initial release with 8 and 12-bit support. Currently configured with...

• 8-bit sample resolution
• 8kHz sample rate and 14kHz playback rate
• Silence attenuation
• 4-second sample buffer