Abstract
This paper comes up with “Hybridized Delta Spectral Cepstral Coefficients” or HDSCC features. The HDSCC feature is obtained by processing concatenated Mel Filter Cepstral Coefficients and Delta Spectral Cepstral Coefficients (DSCC) with Quantile based Dynamic Cepstral Normalization technique (QCN). The strength of proposed HDSCC feature set has been evaluated for Hindi Vowel classification in noisy environments. The results have been compared with that obtained with MFCC and QCN-MFCC features. HDSCC feature set outweighs classification accuracy compared to that obtained with MFCC features and QCN-MFCC features in various operating conditions. Quantitatively for clean database results revealed an improvement of 20.33 and 15.93% as compared to that obtained with MFCC and QCN-MFCC features respectively for context independent (CI) cases. For context dependent (CD) cases an improvement of 18.59 and 11.69% has been detected for MFCC features and QCN-MFCC features respectively. For noisy database also HDSCC features demonstrate better efficiency than baseline features. The experiments for the classification of Hindi Vowels, revealed as high as 97.87 and 94.85% average % Vowel Classification Accuracy (% VCA) in noisy environments for CD cases and CI cases respectively. The maximum comparative improvement in % VCA for CI cases is noted down as 29.45 and 25.88% over MFCC and QCN-MFCC features respectively while the same for CD cases has been detected as 27.09 and 24.12% over MFCC and QCN-MFCC features respectively. All results are obtained on Matlab R2009b with HMM classifier.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Biswas A, Sahu P, Bhowmick A, Chandra M (2015) AAM based features for multiple camera visual speech recognition in car environment. Procedia Comput Sci 57:614–621
Kaynak MN, Zhi Q, Cheok AD, Sengupta K, Jian Z, Chung KC (2004) Lip geometric features for human computer interaction using bimodal speech recognition: comparison and analysis. Speech Commun 43(1):1–16
Potamianos G, Neti C, Luettin J, Matthews I (2004) Audio visual automatic speech recognition: an overview. Issues Visual Audio-Visual Speech Process 22:23
Biawas A, Sahu P, Chandra M (2014) Admissible wavelet packet features based on human inner ear frequency response for Hindi consonant recognition. Comput Electr Eng 40(4):1111–1122
Farooq O, Datta S, Shrotriya MC (2010) Wavelet sub-band based temporal features for robust Hindi phoneme recognition. Int J Wavelets Multiresolut Inf Process 8:847–859
Wong E, Sridharan S (2001) Comparison of linear prediction cepstrum coefficients and mel-frequency cepstrum coefficients for language identification. In: Proceedings of IEEE internatinal symposium on intelligent multimedia, video and speech processing, pp 95–98
Kim C, Stern RM (2016) Power-normalized cepstral coefficients (PNCC) for robust speech recognition. IEEE/ACM Trans Audio Speech Lang Process 24(7):1315–1329
Kumar K, Kim C, Stern RM (2011) Delta-spectral cepstral coefficients for robust speech recognition. In: ICASSP 2011 IEEE international conference on acoustics, speech and signal processing, ICASSP-2011, pp 4784–4787
Bořil H, Hansen JHL (2010) Unsupervised equalization of Lombard effect for speech recognition in noisy adverse environments. IEEE Trans Audio Speech Lang Process 18(6):1379–1393
Mishra S, Chandra M (2016) Hindi vowel classification using QCN-PNCC features. Indian J Sci Technol 9(38). https://doi.org/10.17485/ijst/2016/v9i38/102972
Davis SB, Mermelstein P (1980) Comparison of parametric representations for monosyllabic word recognition in continuously spoken sentences. IEEE Trans Acoust Speech Signal Process—ASSP 28:357–366
JainP, Hermansky H (2001) Improved mean and variance normalization for robust speech recognition. In: Proceesings of the IEEE international conference on acoustics, speech, and signal processing
Harvilla MJ, Stern RM (2012) Histogram-based subband power warping and spectral averaging for robust speech recognition under matched and multistyle training. In: Proceesings of the IEEE international conference on acoustics, speech, and signal processing, pp 4697–4700
Samudravijaya K, Rao PVS, Agrawal SS (2002) Hindi speech database. In: International conference on spoken language processing (ICSLP00), Beijing, pp 456–459
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Shipra, Chandra, M. (2019). Effect of Processing Combined MFCC and DSCC Features with QCN for Hindi Vowel Classification in Noisy Environments. In: Nath, V., Mandal, J. (eds) Nanoelectronics, Circuits and Communication Systems . Lecture Notes in Electrical Engineering, vol 511. Springer, Singapore. https://doi.org/10.1007/978-981-13-0776-8_3
Download citation
DOI: https://doi.org/10.1007/978-981-13-0776-8_3
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-13-0775-1
Online ISBN: 978-981-13-0776-8
eBook Packages: EngineeringEngineering (R0)