This demo (or actually 2) streams adpcm-encoded sampled music while also displaying the lyrics. "Karaoke style" if you will. The adpcm-encoding is not really needed because the SD-card storage is enormous, but it's a nice gimmick that reduces the sound data size by a factor of 4, and showing that the X16 is fast enough to decode a high quality sound stream. The sound is mono 16 bit at 20 kHz.
A Python script is used to generate the lyric sync timestamps based off (hand-timed) timestamps in the lyric source file. You also need the sox and adpcm-xq tools to generate the sound data files from the original mp3s.
The songs are two songs by Chvrches that they made for the computer games Mirror's Edge Catalyst and Death Stranding.
