Abstract
Recently, the extraction of clean speech from the noisy speech has gained significance in designing the smart hearing aid by using deep neural network-based speech enhancement techniques. In the conventional denoising process, irrespective of the type of the noise, noise was totally removed from the noisy speech signal. But some of the noises such as alert sounds, siren, fire alarm, baby cry are essential and also known as desired noise. Due to this problem, people with hearing issues were troubled to hear these essential noises. The proposed work provides a solution for this problem by classifying the noises into desired and undesired noise using deep learning technique with convolutional neural network. The desired noise is fused with the clean speech and provides the desired noisy speech signal. Speech enhancement is carried out using a suitable Deep Neural network while classification of noise is carried out using Convolutional Neural Network. Experimental setup for extraction of clean speech segments from noisy version, classification of noise based on its characteristics and weighted fusion desired noise with the enhanced speech have been presented in this proposed work. By using deep learning techniques, the better accuracy has been achieved.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
fxprosound audio design, Traffic Roadsounds, Jul. 2009. [Online]. Available: https://www.freesound.org/s/75375/
Raghuvanshi, S., Goyal, S.: Development of digital signal processing platform for digital hearing aid. Int. J. Adv. Res. Electrical Electronics Instrumentation Eng. (An ISO 3297: 2007 Certified Organization) 3(2), February 2014
Gerkmann, T., Hendriks, R.C.: Noise power estimation based on the probability of speech presence. In: Proc IEEE Workshop Appl. Signal Process. Audio Acoust, pp. 145–148. New Paltz, NY, USA (2011)
Breithaupt, C., Krawczyk, M., Martin, R.: Parameterized MMSE spectral magnitude estimation for the enhancement of noisy speech. In :Proc Speech Signal Process, pp. 4037–4040. , Las Vegas, NV, USA, Apr, IEEE Int. Conf. Acoust. (2008)
Barchiesi, D., Giannoulis, D., Stowell, D., Plumbley, M.D.: Acoustic scene classification: classifying environments from the sounds they produce. IEEE Signal Process. Mag. 32(3), 16–34 (2015). https://doi.org/10.1109/MSP.2014.2326181
Hinton, G.E., Srivastava, N., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.R.: Improving neural networks by preventing co-adaptation of feature detectors. Arxiv (2012) (Online). Available: http://arxiv.org/abs/1207.0580
Hu, G.: 100 nonspeech environmental sounds (2004) (Online). Available: http://www.cse.ohiostate.edu/pnl/corpus/HuCorpus. html
Piczak. K.J.: Dataset for environmental sound classification. In: 23rd ACM International Conference on Multimedia, Brisbane, Australia, Oct. 2015, pp. 1015–1018
Garofolo, J.S.: Getting started with the DARPA TIMIT CD-ROM: an acoustic phonetic continuous speech database, NIST Tech. Rep (1988)
Khandelwal, R., Narayananm S., Lim L.: Emergency alert service (2006) (Online). Available: https://patents.google.com/patent/ US7119675B2/en
Fu, S.W., Tsao, Y., Lu, X., Kawaim H.: Raw waveform-based speech enhancement by fully convolutional networks. ArXiv (2017) (Online). Available: https://arxiv.org/abs/1703.02205
Park, S.R., Lee, J.: A fully convolutional neural network for speech enhancement. Arxiv (2016) (Online). Available: https://arxiv.org/abs/1609.07132
Abdel-Hamid, O., Jiang, H.: Fast speaker adaptation of hybrid NN/HMM model for speech recognition based on discriminative learning of speaker code. In: Proc. ICASSP, pp. 7942–7946 (2013)
Erhan, D., Bengio, Y., Courville, A., Manzagol, P.-A., Vincent, P.: Why does unsupervised pre-training help deep learning? J. Mach. Learn. Res. 11, 625–660 (2010)
Lim, J.S., Oppenheim, A.V.: Enhancement and bandwidth compression of noisy speech. Proc. IEEE 67(12), 1586–1604 (1979)
Rehr, R., Gerkmann, T.: On the importance of Super-Gaussian speech priors for machine-learning based speech enhancement. IEEE/ACM Trans. Audio Speech Language Process. 26(2), Feb. 2018
Ephraim, Y., Malah, D.: Speech enhancement using a minimum mean-square error log-spectral amplitude estimator. IEEE Trans. Acoust. Speech Signal Process ASSP 33(2), 443–445 (1985)
Erhan, D., Courville, A., Bengio, Y., Vincent, P.: Why does unsupervised pre-training help deep learning. J. Mach. Learn. Res. 11, 625–660 (2010)
Cohen, I., Berdugo, B.: Speech enhancement for non-stationary noise environments. Signal Process 81(11), 2403–2418 (2001)
Xu, Y., Du, J., Dai, L.-R., Lee, C.-H.: Fellow, IEEE. A Regression approach to speech enhancement based on deep neural networks. IEEE/ACM Trans. Audio Speech Language Process. 23(1), January 2015
Wang, Y., Wang, D.: Boosting classification based speech separation using temporal dynamics. In: INTERSPEECH 2012 ISCA’s 13th Annual Conference Portland, OR, USA, September 9–13, 2012
Steeneken, H.J.M., Geurtsen, F.W.M.: Description of the RSG.10 noise database, TNO Institute for Perception, Soesterberg, The Netherlands, Tech. Rep. IZF 1988–3 (1988)
Piczak, K.J.: Environmental sound classification with convolutional neural networks. In: MLSP, 1–6 (2015)
Ephraim, Y., Malah, D.: Speech enhancement using a minimum-m square error short-time spectral amplitude estimator. IEEE Trans. Acoust. Speech Signal Process 32(6), 1109–1121 (1984)
Dieleman, S., Brakel, P., Schrauwen, B.: Audiobased music classification with a pretrained convolutional network. In: Proceedings of the 12th International Society for Music Information Retrieval (ISMIR) conference, pp. 669–674 (2011)
Xu, Y., Du, J., Dai, L.-R., Lee, C.-H.: An experimental study on speech enhancement based on deep neural networks. IEEE Signal Process. Lett. 21(1), 65–68 (2014)
Hirsch, H.G., Pearce, D.: The AURORA experimental framework for the performance evaluations of speech recognition systems under noisy conditions. In: Proc. ISCA ITRW ASR, pp. 181–188 (2000)
Lee, H., Grosse, R., Ranganath, R., Ng, A.Y.: Unsupervised learning of hierarchical representations with convolutional deep belief networks. Commun. ACM 54(10), 95–103 (2011)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Mohan, V., Shanmugapriya, P., Sharan Jasmine, A. (2022). Extraction of Clean Speech Along with Emphasis on Essential Noise. In: Joshi, A., Mahmud, M., Ragel, R.G., Thakur, N.V. (eds) Information and Communication Technology for Competitive Strategies (ICTCS 2020). Lecture Notes in Networks and Systems, vol 191. Springer, Singapore. https://doi.org/10.1007/978-981-16-0739-4_3
Download citation
DOI: https://doi.org/10.1007/978-981-16-0739-4_3
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-0738-7
Online ISBN: 978-981-16-0739-4
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)