Článek ve sborníku konference

GRÉZL František a ČERNOCKÝ Jan. TRAP-based Techniques for Recognition of Noisy Speech. In: Proc. 10th International Conference on Text Speech and Dialogue (TSD 2007). Berlin: Springer Verlag, 2007, s. 270-277. ISBN 978-3-540-74627-0.
Jazyk publikace:angličtina
Název publikace:TRAP-based Techniques for Recognition of Noisy Speech
Název (cs):Techniky založené na TRAP pro rozpoznávání zašuměné řeči
Strany:270-277
Sborník:Proc. 10th International Conference on Text Speech and Dialogue (TSD 2007)
Konference:International Conference on Text Speech and Dialogue
Řada knih:LNCS
Místo vydání:Berlin, DE
Rok:2007
ISBN:978-3-540-74627-0
Vydavatel:Springer Verlag
URL:http://www.fit.vutbr.cz/research/groups/speech/publi/2007/grezl_trap_noise_tsd_2007.pdf [PDF]
Klíčová slova
TRAP techniques, noisy speech recognition, multistream processing, feature combination
Anotace
článek prezentuje techniky založené na TRAPech pro rozpoznávání zašuměné řeči, ukazuje zlepšení pokud se informace obsažená v signálu rozdělí a zpracuje samoztatně a po té opět spojí.
Abstrakt
This paper presents a systematic study of performance of TempoRAl Patterns (TRAP) based features and their proposed modifications and combinations for speech recognition in noisy environment. The experimental results are obtained on AURORA2 database with clean training data. We observed large dependency of performance of different TRAP modifications on noise level. Earlier proposed TRAP system modifications help in clean conditions but degrade the system performance in presence of noise. The combination techniques on the other hand can bring large improvement in case of weak noise and degrade only slightly for strong noise cases. The vector concatenation combination technique is improving the system performance up to strong noise.
BibTeX:
@INPROCEEDINGS{
   author = {Franti{\v{s}}ek Gr{\'{e}}zl and Jan {\v{C}}ernock{\'{y}}},
   title = {TRAP-based Techniques for  Recognition of Noisy Speech},
   pages = {270--277},
   booktitle = {Proc. 10th International Conference on Text Speech and
	Dialogue (TSD 2007)},
   series = {LNCS},
   year = {2007},
   location = {Berlin, DE},
   publisher = {Springer Verlag},
   ISBN = {978-3-540-74627-0},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php.cs.iso-8859-2?id=8434}
}

Vaše IPv4 adresa: 54.81.254.212
Přepnout na IPv6 spojení

DNSSEC [dnssec]