Abstract
With the technological advancement of smart home devices, the lifestyles of people have been gradually changed. Meanwhile, speaker recognition is available in almost all smart home devices. Currently, the mainstream speaker recognition service is provided by a very deep neural network which trained on the cloud server. However, these deep neural networks are not suitable for deployment and operation on smart home devices. Aiming at this problem, in this paper, we propose a packet bottleneck method to improve SqueezeNet which has been widely used in the speaker recognition task. In the meantime, a lightweight structure named TrimNet has been designed. Besides, a model updating strategy based on transfer learning has been adopted to avoid model deteriorates due to the cold speech. The experimental results demonstrate that the proposed lightweight structure TrimNet is superior to SqueezeNet in classification accuracy, structural parameter quantity, and calculation amount. Moreover, the model updating method can increase the recognition rate of cold speech without damaging the recognition rate of other speakers.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Hansen, J.H.L., Hasan, T.: Speaker recognition by machines and humans: a tutorial review. IEEE Signal Process. Mag. 32(6), 74–99 (2015)
Richards, H., Haynes, R., Kim, Y., Bridle, J.: Generalised discriminative transform via curriculum learning for speaker recognition. In: 2018 IEEE ICASSP, pp. 5324–5328 (2018)
Ghiurcau, M.V., Rusu, C., Astola, J.: A study of the effect of emotional state upon text-independent speaker identification. In: 2011 IEEE International Conference on ICASSP, 2011, pp. 4944–4947 (2011)
Matveev, Y.: The problem of voice template aging in speaker recognition systems. In: Železný, M., Habernal, I., Ronzhin, A. (eds.) SPECOM 2013. LNCS (LNAI), vol. 8113, pp. 345–353. Springer, Cham (2013). https://doi.org/10.1007/978-3-319-01931-4_46
Przybocki, M.A., Martin, A.F., Le, A.N.: Nist speaker recognition evaluations utilizing the mixer corporał 2004, 2005, 2006. IEEE Trans. Audio Speech Lang. Process. 15(7), 1951–1959 (2007)
Wagner, J., Fraga-Silva, T., Josse, Y., Schiller, D., Sei-derer, A., Andre, E.: Infected phonemes: how a cold impairs speech on a phonetic level. In: Proceedings of Interspeech 2017, pp. 3457–3461 (2017)
Bengio, Y.: End-to-end attention-based large vocabulary speech recognition. In: 2016 IEEE Inter- national Conference on ICASSP, 2016, pp. 4945–4949 (2016)
Berry, D.A., Herzel, H., Titze, I.R., Krischer, K.: Interpretation of biomechanical simulations of normal and chaotic vocal fold oscillations with empirical eigenfunctions. J. Acoust. Soc. Am. 95(6), 3595–3604 (1994)
Godino Llorente, J.I., Díazde María, F.: Characterization of healthy and pathological voice through measures based on nonlinear dynamics. IEEE Trans. Audio Speech Lang. Process. 17(6), 1186–1195 (2009)
Hansen, J.H.L., Gavidia Ceballos, L., Kaiser, J.F.: A nonlinear operator-based speech feature analysis method with application to vocal fold pathology assessment. IEEE Trans. Biomed. Eng. 45(3), 300–313 (1998)
Tull, R.G., Rutledge, J.C., Larson, C.R: Cepstral analysis of cold-speech for speaker recognition: a second look. Ph.D. thesis, ASA (1996)
Cole, R.A., Noel, M., Noel, V.: The CSLU speaker recognition corpus. In: Fifth International Conference on Spoken Language Processing (1998)
Beigi, H.: Effects of time lapse on speaker recognition results. In: 2009 16th Inter- national Conference on Digital Signal Processing, pp. 1–6 (2009)
Reynolds, D.A., Rose, R.C., et al.: Robust text-independent speaker identification using gaussian mixture speaker models. IEEE Trans. Speech Audio Process. 3(1), 72–83 (1995)
Ouellet, P.: Front-end factor analysis for speaker verification. IEEE Trans. Speech Audio Process. 19(4), 788–798 (2011)
Senior, I., Lopez-Moreno, A.: Improving DNN speaker independence with i-vector inputs. In: 2014 IEEE International Conference on ICASSP, 2014, pp. 225–229 (2014)
Kenny, P.: Bayesian speaker verification with heavy tailed priors. In: Odyssey 2010, p. 14 (2010)
Rohdin, J., Silnova, A., Diez, M., Plchot, O., Matějka, P., Burget, L.: End-to-end DNN based speaker recognition inspired by i-vector and PLDA. In: 2018 IEEE ICAS-SP, 2018, pp. 4874–4878 (2018)
Yamada, T., Wang, L., Kai, A.: Improvement of distant-talking speaker identification using bottleneck features of DNN. In: Interspeech 2013, pp. 3661–3664 (2013)
Lei, Y., Scheffer, N., Ferrer, L., McLaren, M.: A novel scheme for speaker recognition using a phonetically- aware deep neural network. In: 2014 IEEE International Conference on ICASSP, 2014, pp. 1695–1699 (2014)
Torfi, A., Dawson, J., Nasrabadi, N.M.: Text-independent speaker verification using 3d convolutional neural networks. In: 2018 IEEE ICME, 2018, pp. 1–6 (2018)
Acknowledgement
This paper is supported by the National Natural Science Foundation of China (General Program). Grant No. 61971316.
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Ai, H., Xia, W., Zhang, Q. (2019). Speaker Recognition Based on Lightweight Neural Network for Smart Home Solutions. In: Vaidya, J., Zhang, X., Li, J. (eds) Cyberspace Safety and Security. CSS 2019. Lecture Notes in Computer Science(), vol 11983. Springer, Cham. https://doi.org/10.1007/978-3-030-37352-8_37
Download citation
DOI: https://doi.org/10.1007/978-3-030-37352-8_37
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-37351-1
Online ISBN: 978-3-030-37352-8
eBook Packages: Computer ScienceComputer Science (R0)