• DocumentCode
    2328114
  • Title

    A novel segmentation method of Sound-Packets for Bangla speech signal

  • Author

    Rahaman, M.A.N.R. ; Das, A. ; Nayen, M.Z. ; Rahman, M.S.

  • Author_Institution
    Dept. of Electr. & Electron. Eng., Bangladesh Univ. of Eng. & Technol., Dhaka, Bangladesh
  • fYear
    2010
  • fDate
    18-20 Dec. 2010
  • Firstpage
    510
  • Lastpage
    513
  • Abstract
    This paper describes several Sound-Packet segmentation techniques, which will facilitate Automatic Speech Recognition (ASR) for Bangla speech signal. The approximate duration of a sound-packet has been determined and an envelope-detection method has been presented to determine the end-points of sound-packets. The 1st difference method, based on moving average of 1st difference of the signal, is then introduced. Finally, the Mel-Frequency Cepstral Coefficients (MFCC) are determined and have been applied along with other techniques for successful segmentation of Bangla sound-packets.
  • Keywords
    cepstral analysis; natural language processing; speech recognition; Bangla speech signal; MFCC; automatic speech recognition; envelope-detection method; mel-frequency cepstral coefficients; sound-packet segmentation techniques; Bangla Speech Segmentation; Mel-Frequency Cepstral Coefficients; Moving Average; Sound-Packet; Sound-Unit;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Electrical and Computer Engineering (ICECE), 2010 International Conference on
  • Conference_Location
    Dhaka
  • Print_ISBN
    978-1-4244-6277-3
  • Type

    conf

  • DOI
    10.1109/ICELCE.2010.5700741
  • Filename
    5700741
    • Link To Document
    https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=2328114