4 ms·SampleRNN: An Unconditional End-To-End Neural Audio Generation Model17 points by kebinappies 10y ago