A convolutional network for streaming speech enhancement produces models with configurable latency (12.5-75 ms) via asymmetric temporal padding, reaching PESQ 3.35-3.43 on VoiceBank+DEMAND while matching prior causal SOTA at lowest latency.
Fast Wavenet Generation Algorithm
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
abstract
This paper presents an efficient implementation of the Wavenet generation process called Fast Wavenet. Compared to a naive implementation that has complexity O(2^L) (L denotes the number of layers in the network), our proposed approach removes redundant convolution operations by caching previous calculations, thereby reducing the complexity to O(L) time. Timing experiments show significant advantages of our fast implementation over a naive one. While this method is presented for Wavenet, the same scheme can be applied anytime one wants to perform autoregressive generation or online prediction using a model with dilated convolution layers. The code for our method is publicly available.
verdicts
UNVERDICTED 2representative citing papers
Deep recurrent autoencoders convert images to shortened audio signals that incorporate hearing models, enabling above-chance hand posture discrimination and object reaching after a few hours of training instead of months.
citing papers explorer
-
Latency-Configurable Streaming Speech Enhancement via Asymmetric Temporal Padding
A convolutional network for streaming speech enhancement produces models with configurable latency (12.5-75 ms) via asymmetric temporal padding, reaching PESQ 3.35-3.43 on VoiceBank+DEMAND while matching prior causal SOTA at lowest latency.
-
Autoencoding sensory substitution
Deep recurrent autoencoders convert images to shortened audio signals that incorporate hearing models, enabling above-chance hand posture discrimination and object reaching after a few hours of training instead of months.