Abstract
This paper proposes effective energy feature normalization methods for robust speech recognition in noisy environments. We first develop an energy subtraction method and a modified method for the Log-energy Dynamic Range Normalization (ERN) using inverse function. We then present the hybrid method combining the energy subtraction and the modified ERN. Using Aurora2.0 database for representative evaluations, a significant performance improvement over the ERN method is demonstrated.
Original language | English |
---|---|
Pages (from-to) | 1508-1511 |
Number of pages | 4 |
Journal | IEICE Transactions on Communications |
Volume | E90-B |
Issue number | 6 |
DOIs | |
Publication status | Published - 2007 Jun 1 |
Fingerprint
Keywords
- Energy-subtraction
- Log-energy dynamic range normalization (ERN)
- Modified ERN
- Speech recognition
ASJC Scopus subject areas
- Electrical and Electronic Engineering
- Computer Networks and Communications
- Software
Cite this
Effective energy feature compensation using modified log-energy dynamic range normalization for robust speech recognition. / Lee, Yoonjae; Ko, Hanseok.
In: IEICE Transactions on Communications, Vol. E90-B, No. 6, 01.06.2007, p. 1508-1511.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Effective energy feature compensation using modified log-energy dynamic range normalization for robust speech recognition
AU - Lee, Yoonjae
AU - Ko, Hanseok
PY - 2007/6/1
Y1 - 2007/6/1
N2 - This paper proposes effective energy feature normalization methods for robust speech recognition in noisy environments. We first develop an energy subtraction method and a modified method for the Log-energy Dynamic Range Normalization (ERN) using inverse function. We then present the hybrid method combining the energy subtraction and the modified ERN. Using Aurora2.0 database for representative evaluations, a significant performance improvement over the ERN method is demonstrated.
AB - This paper proposes effective energy feature normalization methods for robust speech recognition in noisy environments. We first develop an energy subtraction method and a modified method for the Log-energy Dynamic Range Normalization (ERN) using inverse function. We then present the hybrid method combining the energy subtraction and the modified ERN. Using Aurora2.0 database for representative evaluations, a significant performance improvement over the ERN method is demonstrated.
KW - Energy-subtraction
KW - Log-energy dynamic range normalization (ERN)
KW - Modified ERN
KW - Speech recognition
UR - http://www.scopus.com/inward/record.url?scp=67650787216&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=67650787216&partnerID=8YFLogxK
U2 - 10.1093/ietcom/e90-b.6.1508
DO - 10.1093/ietcom/e90-b.6.1508
M3 - Article
AN - SCOPUS:67650787216
VL - E90-B
SP - 1508
EP - 1511
JO - IEICE Transactions on Communications
JF - IEICE Transactions on Communications
SN - 0916-8516
IS - 6
ER -