STFT와 RNN을 활용한 화자 인증 모델
Speaker Verification Model Using Short-Time Fourier Transform and Recurrent Neural Network
0
조회
0
다운로드
5
인용
0
참고문헌
키워드
초록
최근 시스템에 음성 인증 기능이 탑재됨에 따라 화자(Speaker)를 정확하게 인증하는 중요성이 높아지고 있다. 이에 따라 다양한 방법으로 화자를 인증하는 모델이 제시되어 왔다. 본 논문에서는 Short-time Fouriertransform(STFT)를 적용한 새로운 화자 인증 모델을 제안한다. 이 모델은 기존의 Mel-Frequency CepstrumCoefficients(MFCC) 추출 방법과 달리 윈도우 함수를 약 66.1% 오버랩하여 화자 인증 시 정확도를 높일 수 있다. 새로운 화자 인증 모델을 제안한다. 이 때, LSTM 셀을 적용한 Recurrent Neural Network(RNN)라는 딥러닝 모델을 사용하여 시변적 특징을 가지는 화자의 음성 특징을 학습하고, 정확도가 92.8%로 기존의 화자 인증 모델보다 5.5% 정확도가 높게 측정되었다.
Abstract
Recently as voice authentication function is installed in the system, it is becoming more important to accuratelyauthenticate speakers. Accordingly, a model for verifying speakers in various ways has been suggested. In this paper, wepropose a new method for verifying speaker verification using a Short-time Fourier Transform(STFT). Unlike the existingMel-Frequency Cepstrum Coefficients(MFCC) extraction method, we used window function with overlap parameter of around66.1%. In this case, the speech characteristics of the speaker with the temporal characteristics are studied using a deeprunning model called RNN (Recurrent Neural Network) with LSTM cell. The accuracy of proposed model is around 92.8%and approximately 5.5% higher than that of the existing speaker certification model.
참고문헌 (23)
- Z. Zhang, "Text-dependent speaker verification", U. S. Patent no. 8, ~ (2012)
- F. Bimbot, "A tutorial on text-independent speaker verification", EURASIP Journal on Advances in Signal Processing (2004)
- J. Hai, "Improved linearpredictive coding method for speech recognition", Fourth International Conference on Information, 3, pp.1614~1618 (2003)
- L. Muda, "Voice recognition algorithms using mel frequency cepstral coefficient (MFCC) and dynamic time warping (DTW) techniques", Journal of Computing, 2(3) (2010)
- H. Hermansky, "Perceptual linear predictive(PLP)analysis of speech", Journal of the Acoustical Society of America, 87, pp.1738~1752 (1990)
- T. Baba, "Time-Frequency Analysis Using Short Time Fourier Transform", Open Acoustics Journal (2012)
- E. Variani, "Deep neural networks for small footprint text-dependent speaker verification", IEEE International Conference on Acoustics, Speech and Signal Processing(ICASSP) IEEE, pp.4052~4056 (2014)
- H. Sak, "Long short-term memory recurrent neural network architectures for large scale acoustic modeling", Fifteenth annual conference of the international speech communication association (2014)
- C. Van Loan, "Computational frameworks for the fast Fourier transform vol. 10" (1992)
- L. Liu, "Effects of phase on the perception of intervocaticstop consonants", Speech Communication, 22, pp.403~417 (1997)
- K. Paliwal, "Usefulness of phase spectrum in human speech perception", EUROSPEECH, pp.21187~22120 (2003)
- L. D. Alsteris, "Importance of window shape for phase-only reconstruction of speech", IEEE International Conference on Acoustics, Speech and Signal Processing(ICASSP) IEEE, pp.573~576 (2004)
- L. D. Alsteris, "Further intelligibility results from human listening tests using the short-time phase spectrum", Speech Communication, 48, pp.727~736 (2006)
- S. Kanai, "Advances in Neural Information Processing Systems", pp.286~296 (2018)
- I. Loshchilov, "Sgdr: Stochastic gradient descent with warm restarts", arXiv preprint arXiv:1608.03983, ~ (2016)
- J. Duchi, "Adaptive subgradient methods for online learning and stochastic optimization", Journal of Machine Learning Research, pp.2121~2159 (2011)
- D. P. Kingma, "Adam : A method for stochastic optimization", arXivpreprint arXiv : 1412. 6980, ~ (2014)
- M. Abadi, "Tensorflow: A system for large-scale machine learning", 12th USENIX Symposium on Operating Systems Design and Implementation, pp.265~283 (2016)
- A. Nagrani, "Voxceleb: a large-scale speaker identification dataset", arXiv preprint arXiv : 1706. 08612, ~ (2017)
- Y. H. Chen, "Locally-connected and convolutional neural networks for small foot print speaker recognition", Sixteenth Annual Conference of the International Speech Communication Association (2015)
- H. Salehghaffari, "Speaker Verification using Convolutional Neural Networks", arXiv, ~ (2018)
- H. Georg, "End-to-end text-dependent speaker verification", IEEE International Conference on Acoustics, Speech and Signal Processing(ICASSP) IEEE, pp.5115~5119 (2016)
- T. Amirsina, "Text-Independent Speaker Verification Using 3D CNN", arXiv, ~ (2018)