Convolutional, Long Short-Term Memory, fully connected Deep Neural Networks

Both Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) have shown improvements over Deep Neural Networks (DNNs) across a wide variety of speech recognition tasks. CNNs, LSTMs and DNNs are complementary in their modeling capabilities, as CNNs are good at reducing frequency variat...

Full description

Saved in:
Bibliographic Details
Main Authors: Sainath, Tara N., Vinyals, Oriol, Senior, Andrew, Sak, Hasim
Format: Conference Proceeding
Language:English
Subjects:
Context
Hidden Markov models
Neural networks
Noise measurement
Speech
Speech recognition
Training
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:Both Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) have shown improvements over Deep Neural Networks (DNNs) across a wide variety of speech recognition tasks. CNNs, LSTMs and DNNs are complementary in their modeling capabilities, as CNNs are good at reducing frequency variations, LSTMs are good at temporal modeling, and DNNs are appropriate for mapping features to a more separable space. In this paper, we take advantage of the complementarity of CNNs, LSTMs and DNNs by combining them into one unified architecture. We explore the proposed architecture, which we call CLDNN, on a variety of large vocabulary tasks, varying from 200 to 2,000 hours. We find that the CLDNN provides a 4-6% relative improvement in WER over an LSTM, the strongest of the three individual models.
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.2015.7178838