Simultaneously Trained NN-based Acoustic Model and NN-based Feature Extractor

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F15%3A43926599" target="_blank" >RIV/49777513:23520/15:43926599 - isvavai.cz</a>
Result on the web
<a href="http://link.springer.com/chapter/10.1007/978-3-319-24033-6_27" target="_blank" >http://link.springer.com/chapter/10.1007/978-3-319-24033-6_27</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1007/978-3-319-24033-6_27" target="_blank" >10.1007/978-3-319-24033-6_27</a>

Result language
angličtina
Original language name
Simultaneously Trained NN-based Acoustic Model and NN-based Feature Extractor
Original language description
This paper demonstrates how standard feature extraction methods such as PLP can be successfully replaced by a neural network and methods such as mean normalization, variance normalization and delta coefficients can be simultaneously utilized in a neural-network-based acoustic model. Our experiments show that this replacement is significantly beneficial. Moreover, in our experiments, also a neural-networkbased voice activity detector was employed and trained simultaneously with a neural-network-based feature extraction and a neural-networkbased acoustic model. The system performance was evaluated on the British English speech corpus WSJCAM0.
Czech name
—
Czech description
—

Project
<a href="/en/project/DF12P01OVV022" target="_blank" >DF12P01OVV022: ASR- and MT-based Access to a Large Archive of Cultural Heritage (AMALACH)</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year
2015
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
Text, Speech, and Dialogue, 18th International Conference, TSD 2015, Pilsen, Czech Republic, September 14-17, 2015. Proceedings
ISBN
978-3-319-24032-9
ISSN
0302-9743
e-ISSN
—
Number of pages
9
Pages from-to
234-242
Publisher name
Springer
Place of publication
Berlin
Event location
Plzeň, Czech Republic
Event date
Sep 14, 2015
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
000365947800027

Similar results(10)