Abstract
In this chapter, we show that deep neural networks jointly learn the feature representation and the classifier. Through many layers of nonlinear processing, DNNs transform the raw input feature to a more invariant and discriminative representation that can be better classified by the log-linear model. In addition, DNNs learn a hierarchy of features. The lower-level features typically catch local patterns. These patterns are very sensitive to changes in the raw feature. The higher-level features, however, are built upon the low-level features and are more abstract and invariant to the variations in the raw feature. We demonstrate that the learned high-level features are robust to speaker and environment variations.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Good raw features still help though since the existing DNN learning algorithms may generate an underperformed system even if a linear transformation such as discrete cosine transformation (DCT) is applied to the log filter-bank features.
- 2.
This behavior can be alleviated by adding small random noises to each training sample dynamically during the training time.
- 3.
Huang et al. [8] also tried some of the variations such as the number of vowels per second and the speaking rate normalized by the average duration of different phonemes. It was reported that no matter which definition is used the WER pattern is very similar.
References
Andreou, A., Kamm, T., Cohen, J.: Experiments in vocal tract normalization. In: Proceedings of the CAIP Workshop: Frontiers in Speech Recognition II (1994)
Davis, S., Mermelstein, P.: Comparison of parametric representations for monosyllabic word recognition in continuously spoken sentences. IEEE Trans. Acoust., Speech Signal Process. 28(4), 357–366 (1980)
Flego, F., Gales, M.J.: Discriminative adaptive training with VTS and JUD. In: Proceedings of the IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU), pp. 170–175 (2009)
Gales, M.J., Woodland, P.: Mean and variance adaptation within the mllr framework. Comput. Speech Lang. 10(4), 249–264 (1996)
Gales, M.J.: Maximum likelihood linear transformations for HMM-based speech recognition. Comput. Speech Lang. 12(2), 75–98 (1998)
Gunawardana, A., Mahajan, M., Acero, A., Platt, J.C.: Hidden conditional random fields for phone classification. In: Proceedings of the Annual Conference of International Speech Communication Association (INTERSPEECH), pp. 1117–1120 (2005)
Hinton, G.E., Srivastava, N., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.R.: Improving neural networks by preventing co-adaptation of feature detectors. arXiv preprint arXiv:1207.0580 (2012)
Huang, Y., Yu, D., Liu, C., Gong, Y.: A comparative analytic study on the gaussian mixture and context dependent deep neural network hidden markov models. In: Proceedings of the Annual Conference of International Speech Communication Association (INTERSPEECH) (2014)
Imagenet. http://www.image-net.org/
Jarrett, K., Kavukcuoglu, K., Ranzato, M., LeCun, Y.: What is the best multi-stage architecture for object recognition? In: Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 2146–2153 (2009)
Kalinli, O., Seltzer, M.L., Droppo, J., Acero, A.: Noise adaptive training for robust automatic speech recognition. IEEE Trans. Audio, Speech Lang. Process. 18(8), 1889–1901 (2010)
Kim, D.Y., Kwan Un, C., Kim, N.S.: Speech recognition in noisy environments using first-order vector Taylor series. Speech Commun. 24(1), 39–49 (1998)
Kumar, N., Andreou, A.G.: Heteroscedastic discriminant analysis and reduced rank HMMs for improved speech recognition. Speech Commun. 26(4), 283–297 (1998)
Li, J., Deng, L., Yu, D., Gong, Y., Acero, A.: High-performance HMM adaptation with joint compensation of additive and convolutive distortions via vector Taylor series. In: Proceedings of the IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU), pp. 65–70 (2007)
Li, J., Deng, L., Yu, D., Gong, Y., Acero, A.: HMM adaptation using a phase-sensitive acoustic distortion model for environment-robust speech recognition. In: Proceedings of the International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 4069–4072 (2008)
Li, J., Yu, D., Huang, J.T., Gong, Y.: Improving wideband speech recognition using mixed-bandwidth training data in CD-DNN-HMM. In: Proceedings of the IEEE Spoken Language Technology Workshop (SLT), pp. 131–136 (2012)
Lowe, D.G.: Object recognition from local scale-invariant features. In: Proceedings of the Seventh IEEE International Conference on Computer vision, vol. 2, IEEE, pp. 1150–1157. (1999)
Mohamed, A.R., Hinton, G., Penn, G.: Understanding how deep belief networks perform acoustic modelling. In: Proceedings of the International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 4273–4276 (2012)
Moreno, P.J., Raj, B., Stern, R.M.: A vector Taylor series approach for environment-independent speech recognition. In: Proceedings of the International Conference on Acoustics, Speech and Signal Processing (ICASSP), vol. 2, pp. 733–736 (1996)
Parihar, N., Picone, J.: Aurora working group: DSR front end LVCSR evaluation AU/384/02. Institute for Signal and Information Process, Mississippi State University, Technical Report (2002)
Povey, D., Kanevsky, D., Kingsbury, B., Ramabhadran, B., Saon, G., Visweswariah, K.: Boosted MMI for model and feature-space discriminative training. In: Proceedings of the International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 4057–4060 (2008)
Povey, D., Kingsbury, B., Mangu, L., Saon, G., Soltau, H., Zweig, G.: fMPE: Discriminatively trained features for speech recognition. In: Proceedings of the International Conference on Acoustics, Speech and Signal Processing (ICASSP), vol. 1, pp. 961–964 (2005)
Povey, D., Woodland, P.C.: Minimum phone error and I-smoothing for improved discriminative training. In: Proceedings of the International Conference on Acoustics, Speech and Signal Processing (ICASSP), vol. 1, pp. I–105 (2002)
Ragni, A., Gales, M.: Derivative kernels for noise robust ASR. In: Proceedings of the IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU), pp. 119–124 (2011)
Ratnaparkhi, A.: A simple introduction to maximum entropy models for natural language processing. IRCS Technical Reports Series, p. 81 (1997)
Sainath, T.N., Kingsbury, B., Mohamed, A.r., Ramabhadran, B.: Learning filter banks within a deep neural network framework. In: Proceedings of the IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU) (2013)
Seide, F., Li, G., Chen, X., Yu, D.: Feature engineering in context-dependent deep neural networks for conversational speech transcription. In: Proceedings of the IEEE Workshop on Automfatic Speech Recognition and Understanding (ASRU), pp. 24–29 (2011)
Seltzer, M., Yu, D., Wang, Y.: An investigation of deep neural networks for noise robust speech recognition. In: Proceedings of the International Conference on Acoustics, Speech and Signal Processing (ICASSP) (2013)
Szegedy, C., Zaremba, W., Sutskever, I., Bruna, J., Erhan, D., Goodfellow, I., Fergus, R.: Intriguing properties of neural networks. arXiv preprint arXiv:1312.6199 (2013)
Wang, Y., Gales, M.J.: Speaker and noise factorization for robust speech recognition. IEEE Trans. Audio, Speech Lang. Process. 20(7), 2149–2158 (2012)
Yu, D., Deng, L., Acero, A.: Hidden conditional random field with distribution constraints for phone classification. In: Proceedings of the Annual Conference of International Speech Communication Association (INTERSPEECH), pp. 676–679 (2009)
Yu, D., Ju, Y.C., Wang, Y.Y., Zweig, G., Acero, A.: Automated directory assistance system-from theory to practice. In: Proceedings of the Annual Conference of International Speech Communication Association (INTERSPEECH), pp. 2709–2712 (2007)
Yu, D., Seltzer, M.L., Li, J., Huang, J.T., Seide, F.: Feature learning in deep neural networks—studies on speech recognition tasks. In: Proceedings of the ICLR (2013)
Zeiler, M.D., Fergus, R.: Visualizing and understanding convolutional neural networks. arXiv preprint arXiv:1311.2901 (2013)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2015 Springer-Verlag London
About this chapter
Cite this chapter
Yu, D., Deng, L. (2015). Feature Representation Learning in Deep Neural Networks. In: Automatic Speech Recognition. Signals and Communication Technology. Springer, London. https://doi.org/10.1007/978-1-4471-5779-3_9
Download citation
DOI: https://doi.org/10.1007/978-1-4471-5779-3_9
Published:
Publisher Name: Springer, London
Print ISBN: 978-1-4471-5778-6
Online ISBN: 978-1-4471-5779-3
eBook Packages: EngineeringEngineering (R0)