원문정보
초록
영어
We have already proposed a new feature extraction method based on higher-order local auto-correlation and Fisher weight map (FWM) at Interspeech2006. This paper shows effectiveness
of the proposed FWM in speaker dependent and speaker independent phoneme recognition. Widely used MFCC features lack temporal dynamics. To solve this problem, local auto-correlation features are computed and accumulated by weighting high scores on the discriminative areas. This score map is called Fisher weight map. From the speaker dependent
phoneme recognition, the proposed FWM showed 79.5% recognition rate, by 5.0 points higher than the result by MFCC. Furhermore by combing FWM with MFCC and ¢MFCC, the recognition rate improved to 88.3%. In the speaker independent phoneme recognition, it showed 84.2% recognition rate, by 11.0 points higher than the result by MFCC. By combining FWM with MFCC and ¢MFCC, the reecognition rate improved to 89.0%.
목차
1. Introduction
2. Extraction flow of geometrical discriminative features
3. Local features and weighted higher order local auto-correlations
3.1 Local features
3.2 Weighted higher order local auto-correlations
4. Fisher weight map
5. Phoneme recognition experiments
5.1 Experimental setup
5.2 Speaker dependent phoneme recognition using single feature
5.3 Speaker dependent phoneme recognition by feature integration
5.4 Speaker independent phoneme recognition using single feature
5.5 Speaker independent phoneme recognition by feature integration
6. Conclusion
References