Fast Wavenet Generation Algorithm

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 cal...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2016-11
Main Authors: Tom Le Paine, Khorrami, Pooya, Chang, Shiyu, Zhang, Yang, Ramachandran, Prajit, Hasegawa-Johnson, Mark A, Huang, Thomas S
Format: Article
Language:English
Subjects:
Algorithms
Caching
Complexity
Convolution
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary: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.
ISSN:2331-8422