Abstract
This paper proposes a robust voice activity detection (VAD) method that operates in the presence of noise. For noise robust VAD, we have already proposed statistical models and a switching Kalman filter (SKF)-based technique. In this paper, we focus on a model re-estimation method using Gaussian pruning with weight normalization. The statistical model for SKF-based VAD is constructed using Gaussian mixture models (GMMs), and consists of pre-trained silence and clean speech GMMs and a sequentially estimated noise GMM. However, the composed model is not optimal in that it does not fully reflect the characteristics of the observed signal. Thus, to ensure the optimality of the composed model, we investigate a method for its re-estimation that reflects the characteristics of the observed signal sequence. Since our VAD method works through the use of frame-wise sequential processing, processing with the smallest latency is very important. In this case, there are insufficient re-training data for a re-estimation of all the Gaussian parameters. To solve this problem, we propose a model re-estimation method that involves the extraction of reliable characteristics using Gaussian pruning with weight normalization. Namely, the proposed method re-estimates the model by pruning non-dominant Gaussian distributions that express the local characteristics of each frame and by normalizing the Gaussian weights of the remaining distributions. In an experiment using a speech corpus for VAD evaluation, CENSREC-1-C, the proposed method significantly improved the VAD performance with compared that of the original SKF-based VAD. This result confirmed that the proposed Gaussian pruning contributes to an improvement in VAD accuracy.
Original language | English |
---|---|
Pages (from-to) | 229-244 |
Number of pages | 16 |
Journal | Speech Communication |
Volume | 54 |
Issue number | 2 |
DOIs | |
Publication status | Published - 2012 Feb |
Externally published | Yes |
Fingerprint
Keywords
- Gaussian pruning
- Gaussian weight normalization
- Posterior probability
- Switching Kalman filter
- Voice activity detection
ASJC Scopus subject areas
- Language and Linguistics
- Linguistics and Language
- Communication
- Software
- Computer Vision and Pattern Recognition
- Computer Science Applications
- Modelling and Simulation
Cite this
Frame-wise model re-estimation method based on Gaussian pruning with weight normalization for noise robust voice activity detection. / Fujimoto, Masakiyo; Watanabe, Shinji; Nakatani, Tomohiro.
In: Speech Communication, Vol. 54, No. 2, 02.2012, p. 229-244.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Frame-wise model re-estimation method based on Gaussian pruning with weight normalization for noise robust voice activity detection
AU - Fujimoto, Masakiyo
AU - Watanabe, Shinji
AU - Nakatani, Tomohiro
PY - 2012/2
Y1 - 2012/2
N2 - This paper proposes a robust voice activity detection (VAD) method that operates in the presence of noise. For noise robust VAD, we have already proposed statistical models and a switching Kalman filter (SKF)-based technique. In this paper, we focus on a model re-estimation method using Gaussian pruning with weight normalization. The statistical model for SKF-based VAD is constructed using Gaussian mixture models (GMMs), and consists of pre-trained silence and clean speech GMMs and a sequentially estimated noise GMM. However, the composed model is not optimal in that it does not fully reflect the characteristics of the observed signal. Thus, to ensure the optimality of the composed model, we investigate a method for its re-estimation that reflects the characteristics of the observed signal sequence. Since our VAD method works through the use of frame-wise sequential processing, processing with the smallest latency is very important. In this case, there are insufficient re-training data for a re-estimation of all the Gaussian parameters. To solve this problem, we propose a model re-estimation method that involves the extraction of reliable characteristics using Gaussian pruning with weight normalization. Namely, the proposed method re-estimates the model by pruning non-dominant Gaussian distributions that express the local characteristics of each frame and by normalizing the Gaussian weights of the remaining distributions. In an experiment using a speech corpus for VAD evaluation, CENSREC-1-C, the proposed method significantly improved the VAD performance with compared that of the original SKF-based VAD. This result confirmed that the proposed Gaussian pruning contributes to an improvement in VAD accuracy.
AB - This paper proposes a robust voice activity detection (VAD) method that operates in the presence of noise. For noise robust VAD, we have already proposed statistical models and a switching Kalman filter (SKF)-based technique. In this paper, we focus on a model re-estimation method using Gaussian pruning with weight normalization. The statistical model for SKF-based VAD is constructed using Gaussian mixture models (GMMs), and consists of pre-trained silence and clean speech GMMs and a sequentially estimated noise GMM. However, the composed model is not optimal in that it does not fully reflect the characteristics of the observed signal. Thus, to ensure the optimality of the composed model, we investigate a method for its re-estimation that reflects the characteristics of the observed signal sequence. Since our VAD method works through the use of frame-wise sequential processing, processing with the smallest latency is very important. In this case, there are insufficient re-training data for a re-estimation of all the Gaussian parameters. To solve this problem, we propose a model re-estimation method that involves the extraction of reliable characteristics using Gaussian pruning with weight normalization. Namely, the proposed method re-estimates the model by pruning non-dominant Gaussian distributions that express the local characteristics of each frame and by normalizing the Gaussian weights of the remaining distributions. In an experiment using a speech corpus for VAD evaluation, CENSREC-1-C, the proposed method significantly improved the VAD performance with compared that of the original SKF-based VAD. This result confirmed that the proposed Gaussian pruning contributes to an improvement in VAD accuracy.
KW - Gaussian pruning
KW - Gaussian weight normalization
KW - Posterior probability
KW - Switching Kalman filter
KW - Voice activity detection
UR - http://www.scopus.com/inward/record.url?scp=80055089790&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=80055089790&partnerID=8YFLogxK
U2 - 10.1016/j.specom.2011.08.005
DO - 10.1016/j.specom.2011.08.005
M3 - Article
AN - SCOPUS:80055089790
VL - 54
SP - 229
EP - 244
JO - Speech Communication
JF - Speech Communication
SN - 0167-6393
IS - 2
ER -