Abstract
This paper demonstrates how standard feature extraction methods such as PLP can be successfully replaced by a neural network and methods such as mean normalization, variance normalization and delta coefficients can be simultaneously utilized in a neural-network-based acoustic model. Our experiments show that this replacement is significantly beneficial. Moreover, in our experiments, also a neural-network-based voice activity detector was employed and trained simultaneously with a neural-network-based feature extraction and a neural-network-based acoustic model. The system performance was evaluated on the British English speech corpus WSJCAM0.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Grézl, F., Karafiát, M.: Semi-supervised bootstrapping approach for neural network feature extractor training. In: ASRU, pp. 470–475. IEEE (2013)
Sainath, T.N., Kingsbury, B., Mohamed, A.R, Ramabhadran, B.: Learning filter banks within a deep neural network framework. In: ASRU, pp. 297–302. IEEE (2013)
Narayanan, A., Wang, D.: Ideal ratio mask estimation using deep neural networks for robust speech recognition. In: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 7092–7096, May 2013
Fernandez Astudillo, R., Abad, A., Trancoso, I.: Accounting for the residual uncertainty of multi-layer perceptron based features. In: 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 6859–6863, May 2014
Seps, L., Málek, J., Cerva, P., Nouza, J.: Investigation of deep neural networks for robust recognition of nonlinearly distorted speech. In: INTERSPEECH, pp. 363–367 (2014)
Sainath, T.N., Peddinti, V., Kingsbury, B., Fousek, P., Ramabhadran, B., Nahamoo, D.: Deep scattering spectra with deep neural networks for LVCSR tasks. In: 15th Annual Conference of the International Speech Communication Association, INTERSPEECH 2014, September 14–18, 2014, Singapore, pp. 900–904 (2014)
Tüske, Z., Golik, P., Schlüter, R., Ney, H.: Acoustic modeling with deep neural networks using raw time signal for lvcsr. In: Interspeech, Singapore, pp. 890–894, September 2014
Heřmanský, H.: Perceptual linear predictive (PLP) analysis of speech. J. Acoust. Soc. Am. 57(4), 1738–52 (1990)
Chang, S., Morgan, N.: Robust CNN-based speech recognition with Gabor filter kernels. In: 15th Annual Conference of the International Speech Communication Association, INTERSPEECH 2014, September 14–18, 2014, Singapore, pp. 905–909 (2014)
Robinson, T., Fransen, J., Pye, D., Foote, J., Renals, S.: Wsjcam0: a british english speech corpus for large vocabulary continuous speech recognition. In: Proc. ICASSP 1995, pp. 81–84. IEEE (1995)
Psutka, J., Švec, J., Psutka, J.V., Vaněk, J., Pražák, A., Šmídl, L.: Fast phonetic/lexical searching in the archives of the czech holocaust testimonies: advancing towards the MALACH project visions. In: Sojka, P., Horák, A., Kopeček, I., Pala, K. (eds.) TSD 2010. LNCS, vol. 6231, pp. 385–391. Springer, Heidelberg (2010)
Pražák, A., Psutka, J.V., Psutka, Loose, Z.: Towards live subtitling of TV ice-hockey commentary. In: Cabello, E., Virvou, M., Obaidat, M.S., Ji, H., Nicopolitidis, P., Vergados, D.D. (eds.) SIGMAP, pp. 151–155. SciTePress (2013)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Zelinka, J., Vaněk, J., Müller, L. (2015). Simultaneously Trained NN-Based Acoustic Model and NN-Based Feature Extractor. In: Král, P., Matoušek, V. (eds) Text, Speech, and Dialogue. TSD 2015. Lecture Notes in Computer Science(), vol 9302. Springer, Cham. https://doi.org/10.1007/978-3-319-24033-6_27
Download citation
DOI: https://doi.org/10.1007/978-3-319-24033-6_27
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-24032-9
Online ISBN: 978-3-319-24033-6
eBook Packages: Computer ScienceComputer Science (R0)