엔트로피 차와 신호의 에너지에 기반한 잡음환경에서의 음성검출
Voice Activity Detection Based on Signal Energy and Entropy-difference in Noisy Environments
1
조회
0
다운로드
4
인용
0
참고문헌
키워드
초록
In many areas of speech signal processing such as automatic speech recognition and packet based voice communication technique, VAD (voice activity detection) plays an important role in the performance of the overall system. In this paper, we present a new feature parameter for VAD which is the product of energy of the signal and the difference of two types of entropies. For this end, we first define a Mel filter-bank based entropy and calculate its difference from the conventional entropy in frequency domain. The difference is then multiplied by the spectral energy of the signal to yield the final feature parameter which we call PEED (product of energy and entropy difference). Through experiments, we could verify that the proposed VAD parameter is more efficient than the conventional spectral entropy based parameter in various SNRs and noisy environments.
Abstract
In many areas of speech signal processing such as automatic speech recognition and packet based voice communication technique, VAD (voice activity detection) plays an important role in the performance of the overall system. In this paper, we present a new feature parameter for VAD which is the product of energy of the signal and the difference of two types of entropies. For this end, we first define a Mel filter-bank based entropy and calculate its difference from the conventional entropy in frequency domain. The difference is then multiplied by the spectral energy of the signal to yield the final feature parameter which we call PEED (product of energy and entropy difference). Through experiments, we could verify that the proposed VAD parameter is more efficient than the conventional spectral entropy based parameter in various SNRs and noisy environments.
참고문헌 (11)
- L. Rabiner, "of speech recognitio" (1993), Prentice Hall
- J. Ramírez, "An effective subband OSF-based VAD with noise reduction for robust speech recognition", IEEE Trans. on Speech and Audio Processing, 13(6), pp.1119~1129 (2005)
- Gemello, R, "A modified Ephraim-Malah noise suppression rule for automatic speech recognition", Proc. ICASSP 2004, 1, pp.957~960 (2004)
- P. Renevey, "Entropy based voice activity detection in very noisy conditions", Proc. of Eurospeech, pp.1887~1890 (2001)
- B. F. Wu, "A Noise Estimator with Rapid Adaptation in Variable-Level Noisy Environments", R.O.C. Computational Linguistics Conference, pp.33~38 (2004)
- J. M. Górriz, "An Efficient Bispectrum Phase Entropy- based Algorithm for VAD", Interspeech 2006-ICSLP, pp.17~19 (2006)
- R. R. Venkatesha Prasad, "Voice Activity Detection for VoIP-An Information Theoretic Approach", 49th IEEE Global Telecommunications Conference (2006)
- Shannon, C. E., "A mathematical theory of communication", Bell System Technical Journal, 27, pp.379~423 (1948)
- S. V. Gerven, "A comparative study of speech detection methods", pp.1095~1098 (1997), Eurospeech
- B. F. Wu, "Voice Activity Detection Based on Auto Correlation Function Using Wavelet Transform", Computational Linguistics and Chinese Language Processing, 11(1), pp.87~100 (2006)
- S. S. Stevens, "A Scale for the Measurement of the Psychological Magnitude Pitch", The Journal of the Acoustical Society of America, 8(3), pp.185~190 (1937)