Efficient WaveRNN: Optimizing Nonlinearities
Implementing efficient nonlinearities for WaveRNN CPU inference is tricky but critical.
Andrew Gibiansky
Efficient WaveRNN: Block Sparsity
WaveRNN inference can be accelerated by using block-sparse weight matrices combined with specialized block-sparse matrix-vector multiplication kernels.
Efficient WaveRNN: Optimizing Arithmetic
WaveRNN inference can be made dramatically faster with SIMD inference kernels and int16 quantization.
Efficient WaveRNN: Autoregressive Inference
If implemented in Python and Pytorch, WaveRNN inference is too slow, but we can make it faster with several simple optimizations.
Efficient WaveRNN: Reducing Quantization Noise
Discretizing audio samples to 8 bits each during WaveRNN inferences introduces significant quantization noise, which we can reduce using ยต-law companding and pre-emphasis.
Efficient WaveRNN: Intro
WaveRNN is an autoregressive neural vocoder, a neural network based process for converting low-dimensional acoustic features into an audio waveform by predicting the next sample in a stream of samples. Specialized compute kernels are necessary to make WaveRNN inference fast.