Abstract
One of the most effective approaches to noise robust speech recognition is to remove the noise effect directly from corrupted MFCC vectors. However, VTS enhancement, which is a typical method for performing MFCC enhancement, provides limited improvement when the noise is highly non-stationary. This is because the VTS enhancement method cannot use a time-varying noise model to keep the computational cost at an acceptable level. This paper proposes a method that can enhance MFCC vectors and their dynamic parameters by using noise estimates that change on a frame-by-frame basis at a practical computational cost. The proposed method employs stereo data-based feature mapping like the well known SPLICE algorithm. The novelty of the proposed method lies in that it uses the joint space spanned by a concatenated vector of corrupted and noise features. It is also proposed to use linear discriminant analysis to effectively reduce the dimensionality of the joint space. The proposed method achieves 19.1% and 8.3% relative error reduction from the SPLICE and noise-mean normalized SPLICE algorithms, respectively.
| Original language | English |
|---|---|
| Title of host publication | 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings |
| Pages | 4109-4112 |
| Number of pages | 4 |
| DOIs | |
| Publication status | Published - 2012 |
| Externally published | Yes |
| Event | 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Kyoto Duration: 2012 Mar 25 → 2012 Mar 30 |
Other
| Other | 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 |
|---|---|
| City | Kyoto |
| Period | 12/3/25 → 12/3/30 |
Fingerprint
Keywords
- Noise robust ASR
- non-stationary noise
- SPLICE
ASJC Scopus subject areas
- Signal Processing
- Software
- Electrical and Electronic Engineering
Cite this
MFCC enhancement using joint corrupted and noise feature space for highly non-stationary noise environments. / Suzuki, Masayuki; Yoshioka, Takuya; Watanabe, Shinji; Minematsu, Nobuaki; Hirose, Keikichi.
2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings. 2012. p. 4109-4112 6288822.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - MFCC enhancement using joint corrupted and noise feature space for highly non-stationary noise environments
AU - Suzuki, Masayuki
AU - Yoshioka, Takuya
AU - Watanabe, Shinji
AU - Minematsu, Nobuaki
AU - Hirose, Keikichi
PY - 2012
Y1 - 2012
N2 - One of the most effective approaches to noise robust speech recognition is to remove the noise effect directly from corrupted MFCC vectors. However, VTS enhancement, which is a typical method for performing MFCC enhancement, provides limited improvement when the noise is highly non-stationary. This is because the VTS enhancement method cannot use a time-varying noise model to keep the computational cost at an acceptable level. This paper proposes a method that can enhance MFCC vectors and their dynamic parameters by using noise estimates that change on a frame-by-frame basis at a practical computational cost. The proposed method employs stereo data-based feature mapping like the well known SPLICE algorithm. The novelty of the proposed method lies in that it uses the joint space spanned by a concatenated vector of corrupted and noise features. It is also proposed to use linear discriminant analysis to effectively reduce the dimensionality of the joint space. The proposed method achieves 19.1% and 8.3% relative error reduction from the SPLICE and noise-mean normalized SPLICE algorithms, respectively.
AB - One of the most effective approaches to noise robust speech recognition is to remove the noise effect directly from corrupted MFCC vectors. However, VTS enhancement, which is a typical method for performing MFCC enhancement, provides limited improvement when the noise is highly non-stationary. This is because the VTS enhancement method cannot use a time-varying noise model to keep the computational cost at an acceptable level. This paper proposes a method that can enhance MFCC vectors and their dynamic parameters by using noise estimates that change on a frame-by-frame basis at a practical computational cost. The proposed method employs stereo data-based feature mapping like the well known SPLICE algorithm. The novelty of the proposed method lies in that it uses the joint space spanned by a concatenated vector of corrupted and noise features. It is also proposed to use linear discriminant analysis to effectively reduce the dimensionality of the joint space. The proposed method achieves 19.1% and 8.3% relative error reduction from the SPLICE and noise-mean normalized SPLICE algorithms, respectively.
KW - Noise robust ASR
KW - non-stationary noise
KW - SPLICE
UR - http://www.scopus.com/inward/record.url?scp=84867614789&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84867614789&partnerID=8YFLogxK
U2 - 10.1109/ICASSP.2012.6288822
DO - 10.1109/ICASSP.2012.6288822
M3 - Conference contribution
SN - 9781467300469
SP - 4109
EP - 4112
BT - 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings
ER -