Current Trends in Signal Processing

Perceptual Features Based Continuous Speech Recognition in Additive Noise Environment Using Various Modelling Techniques

  1. A. Revathi ✉
  2. Y. Venkataraman

Abstract

The main objective of this paper is to discuss the effectiveness of Mel frequency perceptual features and the noise reduction technique in evaluating the performance of multi speaker independent continuous speech recognition system in additive noise environment by using various modelling techniques. The proposed perceptual features are captured and trained using clustering technique, GMM, continuous density HMM and back propagation neural networks. Speech recognition system is evaluated on clean and noisy test speeches by using minimum distance criterion, maximum log likelihood criterion and minimum mean squared error criterion for these speech modelling techniques. Performance of these features is tested on speeches randomly chosen from “TIMIT” speech corpus. This algorithm provides 98.3, 90.8, 99.3 and 97.3% as accuracy for clean speech recognition system evaluated on GMM models and continuous density HMM models, Clustering models and neural network models respectively. Evaluation is done for 300 test speeches. System is also tested on noisy test speech by considering RLS and combination of RLS & wavelets as additional pre- processing techniques and the performance is found to be better than the testing without additional pre- processing. Noises such as babble, M109, white, destroyer engine, paper rustle, F16, factory and destroyerops are considered in this work and these are added to the test speeches at various levels and performance of the system is evaluated for 300 test speeches. Among the various modelling techniques, clustering technique and continuous density HMM technique are performing better for clean speech recognition and noisy speech recognition respectively. GMM and continuous density HMM along with additional pre-processing technique are performing better for noisy speech recognition even if the noise energy is greater than the signal energy.

References (22)

  1. Hynek Hermansky, Kazuhiro Tsuga, Shozo Makino and Hisashi Wakita. Perceptually based processing in automatic speech recognition. Proc. IEEE int. conf. on Acoustics, speech and signal processing . Tokyo. April 1986. 11. 1971-1974 p.
  2. Rabiner.L.&Juang B.H. Hidden Markov models for Speech Recognition. Proc. Techno metrics, August 1991. 33. (3). 251– 272 p. Current Trends in Signal Processing Volume 2 (1-3) Compilation 2012 __________________________________________________________________________________________ ISSN:2277–6176© STM Journals 2012. All Rights Reserved Page 80
  3. HynekHermansky, Nelson Margon, ArunaBayya and Phil Kohn. The challenge of Inverse E: The RASTA PLP method.Proc. Twenty fifth IEEE Asilomar conf. on signals, systems and computers. Pacific Grove, CA, USA. November 1991.2. 800–804 p.
  4. HynekHermansky and Nelson Morgan. RASTA processing of speech. IEEE transactions on speech and audio processing. 1994. 2 (4). 578–589p.
  5. Yuval Cohen and AoramErell. IEEE Transactions on Speech and Audio Processing. 1997. 5(2). 141–148p.
  6. Carlos Lima, Luis. B .Almeida and Joao L.Mohreiro. Robust feature extraction for speech recognition in noisy environment. Proc. Int. conf. Signal Processing . Beijing.August 2002. (6).
  7. Weaam, Alkhaldi, WaleedFakhir and NadduHamdy. Automatic speech/speaker Recognition in noisy environments using wavelet transforms. Proc. 45 th Midwest symposium on circuits and systems. Oklahoma. August 2002. 463–466p.
  8. Mosakiyo Fujimoto and Yasuo Ariki. Robust speech recognition in additive and channel noise environments using GMM and EM algorithm . Proceedings of Int. Conf. on Acoustics, Speech and Signal Processing. May 2004. 941–944p.
  9. AkshiyoK.Swain and Waleed Abdullah. Estimation of LPC parameters of speech signals in noisy environments. Proceedings of TENCON. Thailand. November 2004. 139–142p.
  10. Jose Luis Oropezo Radriguez, SergoiSaurez Guerra and Luis Pastor. Using adaptive filter to increase automatic speech recognition rate in a digit corpus.Progress in pattern Recognition, Image analysis and Applications , Lecture notes in computer science, 200 7, 4756, 78– 87p.
  11. Jose Luis Oropezo Radriguez and Sergoi Saurez Guerra. Advances in Artificial Intelligence. Lecture notes in Computer Science. 2007. 4827. 1015–1024p.
  12. Azmi M.M. and Tolba H. Syllable based automatic Arabic speech recognition in noisy envir onment. Proc. Int. conf. on Audio.language and Image Processing . Shanghai. July 2008. 1436–1441p.
  13. Masakiyo Fujimoto, Kentaro Ishizuka, Tomohiro Nakatani. Study of Integration of Statistical Model -Based Voice Activity Detection and Noise Suppression. Proc. Annual Conf. of INTERSPEECH. Brisbane . Australia. September 2008.2008–2011p.
  14. Xiaokin Li and Yunben Deng. Combining speech energy and edge information for efficient voice activity detection in noisy environments. Proc. 19 th Int.conf. on pattern recognition. Tamba. FL. December 2008. 1–4p.
  15. A. Revathi, R. Ganabathy, Y.Venkata- ramani. Int.Journal of Computer Science and Information Technology. 2009. 1(2). 30–42p. Current Trends in Signal Processing Volume 2 (1-3) Compilation 2012 __________________________________________________________________________________________ ISSN:2277–6176© STM Journals 2012. All Rights Reserved Page 81
  16. A.Revathi,Y.Venkataramani, Perceptual Features based Isolated Digit and Continuous Speech Recognition Using Iterative Clustering Approach. Proc. IEEE International Conference on Networking and Communication . Chennai. December 2009.155–160p.
  17. SerajulHaque, Roberto Togoui,Anthony Zakrich.Int. Journal on Speech Communication. 2009. 1(51). 58–75p.
  18. Tomas Tebem, Werner Verhelst, Francois Capman and FredreicBeangenlie, Improved speech recognition in noisy environments by throat microphone for voice activity detection.Proceedings of 18 th Europeon signal processing conference. August 2010. 1978–1982p.
  19. P.Phani Kumar, K.S.N.Vardhan and K.Sriramakrishna.Int. Journal on computer science and communication. 2010. 1 (2). 41–45p.
  20. RashmiMakhaijani, UrmilaShrawardkar, V.M.Thakre.Int. Journal on Engineering Science and Technology. 2011. 3, (2), 1764–1769p.
  21. A.Revathi and Y.Venkataramani,Speaker independent continuous speech and isolated digits recognition using VQ and HMM.Proc. IEEE Int. Conf. on Communication and Signal processing. Calicut. February 2011, 198–202p.
  22. Rabiner.L., Juang B.H. Fundamentals of speech recognition . Prentice Hall . NJ. 1993. • Carbon Nanotubes • Graphenes • Nanodiamonds • Nano Wires • Nano Matals • Nano Plates • Oxide/ Nan oxide Nano Ceramics • Fullerenes • Quantum Dots • PEG Derivatives • Tectomers • Phosphonic Acid Derivatives www.reinste.com NANO PRODUCTS FOR INDUSTRIES • Plastics & Polymers • Detergents & Sanitizers • Construction • Adhesives & Sealants • Oil, Lubricants, Petroleum & Energy • Paint & Coatings • Rubber & Tyres • Pharmaceuticals & Biotechnology • Defense & R&D • Textiles/Yarns/Fabrics/Footwears Designed to deliver the purest…. Discovering the power of 10 M-9 Nano Ventures

Read more of it

Subscribe to one journal or build your own list of several. One invoice, one payment, access for a full year.

Subscribe to journals →

Publish your article

Peer-reviewed journals across every subject, with open access options. Send your manuscript to the right editor today.

Publish with us → Calculate your APC
Support