TY - CONF
T1 - Bandwidth mismatch compensation for robust speech recognition
AU - Liao, Yuan Fu
AU - Lin, Jeng Shien
AU - Tsai, Wei Ho
N1 - Funding Information:
This work was supported in part by National Science Council (NSC), Taiwan, under contract NSC-91-2219-E-027-004 and in part by Ministry of Education (MOE) under contract EX- 91-E-FA06-4-4. The authors also want to thank the Association for Computational Linguistics and Chinese Language Processing (ROCLING), Taiwan for supporting the TCC300 database. 6.
PY - 2003
Y1 - 2003
N2 - In this paper, an iterative bandwidth mismatch compensation (BMC) algorithm is proposed to alleviate the need of multiple pre-trained models for recognizing different bandwidth speech. The BMC uses the concept of the bandwidth extension as similar as in the speech enhancement approaches. However, it aims at directly improving the recognition accuracy instead of speech intelligence or quality and utilizes only recognizer?s hidden Markov models (HMMs) for both bandwidth mismatch compensation and recognition. The BMC first detects the bandwidth of the input speech signal based on a divergence measurement. The HMM/Gaussian mixture model (GMM)- based method is then used to iteratively segment the input speech utterance and compensates the speech features. Experiments on serious bandwidth mismatched conditions, i.e., training on 8 kHz and testing on 4 kHz or 5.5 kHz bandwidth database have verified the effectiveness of the proposed approach.
AB - In this paper, an iterative bandwidth mismatch compensation (BMC) algorithm is proposed to alleviate the need of multiple pre-trained models for recognizing different bandwidth speech. The BMC uses the concept of the bandwidth extension as similar as in the speech enhancement approaches. However, it aims at directly improving the recognition accuracy instead of speech intelligence or quality and utilizes only recognizer?s hidden Markov models (HMMs) for both bandwidth mismatch compensation and recognition. The BMC first detects the bandwidth of the input speech signal based on a divergence measurement. The HMM/Gaussian mixture model (GMM)- based method is then used to iteratively segment the input speech utterance and compensates the speech features. Experiments on serious bandwidth mismatched conditions, i.e., training on 8 kHz and testing on 4 kHz or 5.5 kHz bandwidth database have verified the effectiveness of the proposed approach.
UR - http://www.scopus.com/inward/record.url?scp=85009160436&partnerID=8YFLogxK
M3 - Paper
AN - SCOPUS:85009160436
SP - 3093
EP - 3096
T2 - 8th European Conference on Speech Communication and Technology, EUROSPEECH 2003
Y2 - 1 September 2003 through 4 September 2003
ER -