A Speaker Pruning Method for Real-Time Speaker Identification System
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
It has been known that GMM (Gaussian Mixture Model) based speaker identification systems using ML (Maximum Likelihood) and WMR (Weighting Model Rank) demonstrate very high performances. However, such systems are not so effective under practical environments, in terms of real time processing, because of their high calculation costs. In this paper, we propose a new speaker-pruning algorithm that effectively reduces the calculation cost. In this algorithm, we select 20% of speaker models having higher likelihood with a part of input speech and apply MWMR (Modified Weighted Model Rank) to these selected speaker models to find out identified speaker. To verify the effectiveness of the proposed algorithm, we performed speaker identification experiments using TIMIT database. The proposed method shows more than 60% improvement of reduced processing time than the conventional GMM based system with no pruning, while maintaining the recognition accuracy.
Abstract
It has been known that GMM (Gaussian Mixture Model) based speaker identification systems using ML (Maximum Likelihood) and WMR (Weighting Model Rank) demonstrate very high performances. However, such systems are not so effective under practical environments, in terms of real time processing, because of their high calculation costs. In this paper, we propose a new speaker-pruning algorithm that effectively reduces the calculation cost. In this algorithm, we select 20% of speaker models having higher likelihood with a part of input speech and apply MWMR (Modified Weighted Model Rank) to these selected speaker models to find out identified speaker. To verify the effectiveness of the proposed algorithm, we performed speaker identification experiments using TIMIT database. The proposed method shows more than 60% improvement of reduced processing time than the conventional GMM based system with no pruning, while maintaining the recognition accuracy.
참고문헌 (11)
- S. Furui, "Speaker-dependent feature extraction, recognition and processing techniques", Speech Communication, 10(5-6), pp.505~520 (1991)
- T. Matsui, "Comparison of text independent speaker recognition methods using VQ-distortion and discrete", continuous HMMs, 1992, pp.2, 157~160
- D. A. Reynolds, "Robust text-independent speaker identification using Gaussian mixture speaker models", IEEE Trans. Speech and Audio Processing, 3(1), pp.72~83 (1995)
- D. A. Reynolds, "Speaker identification and verification using Gaussian mixture speaker models", Speech Communication, 17(1-2), pp.91~108 (1995)
- 김민정, "Gaussian Mixture Model과 프레임 단위 유사도 추정을 이용한 유해동영상 필터링 시스템 구현", 한국지능시스템학회 논문지, 23(2), pp.184~189 (2013), 한국지능시스템학회
- K. Markov, "Text-independent speaker identification on TIMIT database", Proceedings of Acoust. Soc. Jap., 1, pp.83~84 (1995)
- K. Markov, "Frame level likelihood normalization for text-independent speaker identification using Gaussian mixture Models", Proceedings of International Conference on Speech and Language Processing, pp.1764~1767 (1996)
- 김민정, "강건한 문맥독립 화자식별을 위한 프레임 선택방법, 복합방법, 수정된 가중모델순위 방법", 한국음향학회지, 21(8), pp.735~743 (2002), 한국음향학회
- M.J. Kim, "Modified weighting model rank method for improving the performance of real-time text-independent speaker recognition system", Proceedings of Acous. Soc. Kor., pp.107~110 (2002)
- V. Zue, "Speech database development at MIT : TIMIT and beyond", Speech Communication, 9(4), pp.351~356 (1990)
- D. A. Reynolds, "Large population speaker identification using clean and telephone speech", IEEE Signal Processing Letters, 2(3), pp.46~48 (1995)