Quartznet: Deep Automatic Speech Recognition with 1D Time-Channel Separable Convolutions

We propose a new end-to-end neural acoustic model for automatic speech recognition. The model is composed of multiple blocks with residual connections between them. Each block consists of one or more modules with 1D time-channel separable convolutional layers, batch normalization, and ReLU layers. I...

Full description

Saved in:
Bibliographic Details
Main Authors: Kriman, Samuel, Beliaev, Stanislav, Ginsburg, Boris, Huang, Jocelyn, Kuchaiev, Oleksii, Lavrukhin, Vitaly, Leary, Ryan, Li, Jason, Zhang, Yang
Format: Conference Proceeding
Language:English
Subjects:
Acoustics
Automatic speech recognition
Conferences
Convolution
convolutional networks
depthwise separable convolution
Speech processing
time-channel separable convolution
Citations: Items that cite this one
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We propose a new end-to-end neural acoustic model for automatic speech recognition. The model is composed of multiple blocks with residual connections between them. Each block consists of one or more modules with 1D time-channel separable convolutional layers, batch normalization, and ReLU layers. It is trained with CTC loss. The proposed network achieves near state-of-the-art accuracy on LibriSpeech and Wall Street Journal, while having fewer parameters than all competing models. We also demonstrate that this model can be effectively fine-tuned on new datasets.
ISSN:2379-190X
DOI:10.1109/ICASSP40776.2020.9053889