Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Adaptation of an EMG-Based Speech Recognizer via Meta-Learning

Abstract: 

In nonacoustic speech recognition based on electromyography, i.e. on electrical muscle activity captured by noninvasive surface electrodes, differences between recording sessions are known to cause deteriorating system accuracy. Efficient adaptation of an existing system to an unseen recording session is therefore imperative for practical usage scenarios. We report on a meta-learning approach to pretrain a deep neural network frontend for a myoelectric speech recognizer in a way that it can be easily adapted to a new session. Fine-tuning this specially pretrained network yields lower Word Error Rates and higher frame accuracies than fine-tuning a conventionally pretrained network, without creating an increased computational burden on a possibly mobile device.

up
0 users have voted:

Paper Details

Authors:
Krsto Proroković, Michael Wand, Tanja Schultz, Jürgen Schmidhuber
Submitted On:
6 December 2019 - 2:25pm
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Krsto Proroković
Paper Code:
1570567443
Document Year:
2019
Cite

Document Files

Adaptation of an EMG-Based Speech Recognizer via Meta-Learning.pdf

(7)

Subscribe

[1] Krsto Proroković, Michael Wand, Tanja Schultz, Jürgen Schmidhuber, "Adaptation of an EMG-Based Speech Recognizer via Meta-Learning", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4954. Accessed: Dec. 12, 2019.
@article{4954-19,
url = {http://sigport.org/4954},
author = {Krsto Proroković; Michael Wand; Tanja Schultz; Jürgen Schmidhuber },
publisher = {IEEE SigPort},
title = {Adaptation of an EMG-Based Speech Recognizer via Meta-Learning},
year = {2019} }
TY - EJOUR
T1 - Adaptation of an EMG-Based Speech Recognizer via Meta-Learning
AU - Krsto Proroković; Michael Wand; Tanja Schultz; Jürgen Schmidhuber
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4954
ER -
Krsto Proroković, Michael Wand, Tanja Schultz, Jürgen Schmidhuber. (2019). Adaptation of an EMG-Based Speech Recognizer via Meta-Learning. IEEE SigPort. http://sigport.org/4954
Krsto Proroković, Michael Wand, Tanja Schultz, Jürgen Schmidhuber, 2019. Adaptation of an EMG-Based Speech Recognizer via Meta-Learning. Available at: http://sigport.org/4954.
Krsto Proroković, Michael Wand, Tanja Schultz, Jürgen Schmidhuber. (2019). "Adaptation of an EMG-Based Speech Recognizer via Meta-Learning." Web.
1. Krsto Proroković, Michael Wand, Tanja Schultz, Jürgen Schmidhuber. Adaptation of an EMG-Based Speech Recognizer via Meta-Learning [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4954