Time-Delay Neural Network with 3 Frequency Bands Based on Voiced Speech Discrimination in Noise

Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 235)


Information on the time variation in a speech signal is significant when training a neural network for the speech signal input. Therefore, this paper proposes a time-delay neural network with 3 frequency bands based on voiced speech discrimination in the condition of background noises. The effectiveness of the proposed network is experimentally confirmed based on measuring the correct discrimination rates for speech degraded by various noises.


Discrimination rate Voiced speech discrimination Time-delay neural network Frequency band Background noise 


  1. 1.
    Juang CF, Chiou CT, Lai CL (2007) Hierarchical singleton-type recurrent neural fuzzy networks for noisy speech recognition. IEEE Trans Neural Netw 18(3):833–843CrossRefGoogle Scholar
  2. 2.
    Knecht WG, Schenkel ME, Moschytz GS (1995) Neural network filters for speech enhancement. IEEE Trans. Speech Audio Process 3(6):433–438Google Scholar
  3. 3.
    Cong L, Asghar S, Cong B (2000) Robust speech recognition using neural networks and hidden Markov models. In: Proceedings of the international on Information technology: coding and computing, pp 350–354Google Scholar
  4. 4.
    Choi JS (2012) Speech processing system using a noise reduction neural network based on FFT spectrums. J Inf Commun Convergence Eng 10(2):162–167CrossRefGoogle Scholar
  5. 5.
    Hampshire JB, Waibel AH (1990) A novel objective function for improved phoneme recognition using time delay neural networks. IEEE Trans Neural Netw 1(2):216–228Google Scholar
  6. 6.
    Choi JS, Park SJ (2007) Speech enhancement system based on auditory system and time-delay neural network. In: 8th international conference on lecture notes in computer science. LNCS, Part II, pp 153–160Google Scholar
  7. 7.
    Peng Y, Xiong H, Guo C, Liu H, Zou J (2010) Research on the algorithm of communication network speech enhancement based on BP neural network. Int Conf Adv Comput Theor Eng 3:V3-559–V3-562Google Scholar
  8. 8.
    Vieira K, Wilamowski B, Kubichek R (1997) Speaker verification for security systems using artificial neural networks. Int Conf Ind Electron Control Instrum 3:1102–1107Google Scholar
  9. 9.
    Rumelhart DE, Hinton GE, Williams RJ (1986) Learning representations by back-propagation errors. Nature 323:533–536CrossRefGoogle Scholar
  10. 10.
    Hirsch H, Pearce D (2000) The AURORA experimental framework for the performance evaluations of speech recognition systems under noisy conditions. In: Proceedings of the ISCA ITRW ASR2000 on automatic speech recognition: challenges for the next millennium, Paris, FranceGoogle Scholar
  11. 11.
    Leonard RG (1984) A database for speaker independent digit recognition. In: IEEE international conference on acoustics, speech, and signal processing, pp 328–331Google Scholar
  12. 12.
    ITU-T (International Telecommunication Union) Recommendation G. 712 (1996) Transmission performance characteristics of pulse code modulation channels, pp 1–31Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2013

Authors and Affiliations

  1. 1.Department of Electronic Engineering, College of EngineeringSilla UniversityBusanKorea

Personalised recommendations