A Comparison of Convolutional Neural Networks for Glottal Closure Instant Detection from Raw Speech
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F21%3A43962806" target="_blank" >RIV/49777513:23520/21:43962806 - isvavai.cz</a>
Result on the web
<a href="https://ieeexplore.ieee.org/document/9413675" target="_blank" >https://ieeexplore.ieee.org/document/9413675</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1109/ICASSP39728.2021.9413675" target="_blank" >10.1109/ICASSP39728.2021.9413675</a>
Alternative languages
Result language
angličtina
Original language name
A Comparison of Convolutional Neural Networks for Glottal Closure Instant Detection from Raw Speech
Original language description
In this paper, we continue to investigate the use of machine learning for the automatic detection of glottal closure instants (GCIs) from raw speech. We compare several deep one-dimensional convolutional neural network architectures on the same data and show that the InceptionV3 model yields the best results on the test set. On publicly available databases, the proposed 1D InceptionV3 outperforms XGBoost, a non-deep machine learning model, as well as other traditional GCI detection algorithms.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
20205 - Automation and control systems
Result continuities
Project
<a href="/en/project/GA19-19324S" target="_blank" >GA19-19324S: Fully Trainable Deep Neural Network Based Czech Text-to-Speech Synthesis</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Others
Publication year
2021
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2021)
ISBN
978-1-72817-605-5
ISSN
1520-6149
e-ISSN
2379-190X
Number of pages
5
Pages from-to
6938-6942
Publisher name
IEEE
Place of publication
New York
Event location
Toronto, ON, Canada
Event date
Jun 6, 2021
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
000704288407043