Loading...

Filter-bank design based on dependencies between frequency components and phoneme characteristics

Mohammadi, S. H ; Sharif University of Technology | 2011

521 Viewed
  1. Type of Document: Article
  2. Publisher: 2011
  3. Abstract:
  4. Mel-frequency Cepstral coefficients are widely used for feature extraction in speech recognition systems. These features use Mel-scaled filters. A new filter-bank based on dependencies between frequency components and phoneme characteristics is proposed. F-ratio and mutual information are used for this purpose. A new filter-bank is designed in which frequency resolution of sub-band filters is inversely proportional to the computed dependency values. These new filterbank is used instead of Mel-scaled filters for feature extraction. A phoneme recognition experiment on FARSDAT Persian language database showed that features extracted using the proposed filter-bank reach higher accuracy (63.92%) compared to Mel-scaled filter-bank (62.37%)
  5. Keywords:
  6. Dependency value ; F-ratio ; FARSDAT ; Frequency components ; Frequency resolutions ; Mel-frequency cepstral coefficients ; Mutual informations ; Persians ; Phoneme recognition ; Speech recognition systems ; Subbands ; Signal processing ; Speech recognition ; Feature extraction
  7. Source: European Signal Processing Conference, 29 August 2011 through 2 September 2011 ; Septembe , 2011 , Pages 2142-2145 ; 22195491 (ISSN)
  8. URL: http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=7074231&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D7074231