A Performance Evaluation of Several Artificial Neural Networks for Mapping Speech Spectrum Parameters

  • Víctor Yeom-Song
  • Marisol Zeledón-Córdoba
  • Marvin Coto-JiménezEmail author
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 1087)


In this work, we compare different neural network architectures, for the task of mapping spectral coefficients of noisy speech signals with those corresponding to natural speech. In previous works on the subject, fully-connected multilayer perception (MLP) networks and recurrent neural networks (LSTM & BLSTM) have been used. Several references report some initial trial and error processes to determine which architecture to use. Finding the best network type and size is of great importance due to the considerable training time required by some models of recurrent networks. In our work, we conducted extensive tests training more than five hundred networks, with several architectures to determine which cases present significant differences. The results show that for this application of neural networks, the architectures with more layers or the greater number of neurons are not the most convenient, both for the time required in their training and for the adjustment achieved. These results depend on the complexity of the task (the signal-to-noise ratio or SNR) and the amount of data available. This exploration can guide the most efficient use of these types of neural networks in future mapping applications, and can help to optimize resources in future studies by reducing computational time and complexity.


Deep learning LSTM Noise Speech enhancement 



This work was supported by the University of Costa Rica (UCR), Project No. 322-B9-105.


  1. 1.
    Abdel-Hamid, O., Mohamed, A.R., Jiang, H., Penn, G.: Applying convolutional neural networks concepts to hybrid NN-HMM model for speech recognition. In: 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 4277–4280. IEEE (2012)Google Scholar
  2. 2.
    Barron, A.R.: Approximation and estimation bounds for artificial neural networks. Mach. Learn. 14(1), 115–133 (1994)zbMATHGoogle Scholar
  3. 3.
    Coto-Jiménez, M.: Robustness of LSTM neural networks for the enhancement of spectral parameters in noisy speech signals. In: Batyrshin, I., Martínez-Villaseñor, M.L., Ponce Espinosa, H.E. (eds.) MICAI 2018. LNCS (LNAI), vol. 11289, pp. 227–238. Springer, Cham (2018). Scholar
  4. 4.
    Coto-Jiménez, M.: Improving post-filtering of artificial speech using pre-trained LSTM neural networks. Biomimetics 4(2), 39 (2019)CrossRefGoogle Scholar
  5. 5.
    Coto-Jiménez, M., Goddard-Close, J.: LSTM deep neural networks postfiltering for improving the quality of synthetic voices. In: Martínez-Trinidad, J.F., Carrasco-Ochoa, J.A., Ayala-Ramírez, V., Olvera-López, J.A., Jiang, X. (eds.) MCPR 2016. LNCS, vol. 9703, pp. 280–289. Springer, Cham (2016). Scholar
  6. 6.
    Coto-Jiménez, M., Goddard-Close, J., Martínez-Licona, F.: Improving automatic speech recognition containing additive noise using deep denoising autoencoders of LSTM networks. In: Ronzhin, A., Potapova, R., Németh, G. (eds.) SPECOM 2016. LNCS (LNAI), vol. 9811, pp. 354–361. Springer, Cham (2016). CrossRefGoogle Scholar
  7. 7.
    Fan, Y., Qian, Y., Xie, F.L., Soong, F.K.: TTS synthesis with bidirectional LSTM based recurrent neural networks. In: Fifteenth Annual Conference of the International Speech Communication Association (2014)Google Scholar
  8. 8.
    Feng, X., Zhang, Y., Glass, J.: Speech feature denoising and dereverberation via deep autoencoders for noisy reverberant speech recognition. In: 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 1759–1763. IEEE (2014)Google Scholar
  9. 9.
    Funahashi, K.I.: On the approximate realization of continuous mappings by neural networks. Neural Netw. 2(3), 183–192 (1989)CrossRefGoogle Scholar
  10. 10.
    Gers, F.A., Schraudolph, N.N., Schmidhuber, J.: Learning precise timing with LSTM recurrent networks. J. Mach. Learn. Res. 3(Aug), 115–143 (2002)MathSciNetzbMATHGoogle Scholar
  11. 11.
    Graves, A., Fernández, S., Schmidhuber, J.: Bidirectional LSTM networks for improved phoneme classification and recognition. In: Duch, W., Kacprzyk, J., Oja, E., Zadrożny, S. (eds.) ICANN 2005. LNCS, vol. 3697, pp. 799–804. Springer, Heidelberg (2005). Scholar
  12. 12.
    Graves, A., Jaitly, N., Mohamed, A.R.: Hybrid speech recognition with deep bidirectional LSTM. In: 2013 IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU), pp. 273–278. IEEE (2013)Google Scholar
  13. 13.
    Hinton, G., et al.: Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups. IEEE Signal Process. Mag. 29(6), 82–97 (2012)CrossRefGoogle Scholar
  14. 14.
    Huang, J., Kingsbury, B.: Audio-visual deep learning for noise robust speech recognition. In: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 7596–7599. IEEE (2013)Google Scholar
  15. 15.
    Ishii, T., Komiyama, H., Shinozaki, T., Horiuchi, Y., Kuroiwa, S.: Reverberant speech recognition based on denoising autoencoder. In: Interspeech, pp. 3512–3516 (2013)Google Scholar
  16. 16.
    Kominek, J., Black, A.W.: The CMU arctic speech databases. In: Fifth ISCA Workshop on Speech Synthesis (2004)Google Scholar
  17. 17.
    Kumar, A., Florencio, D.: Speech enhancement in multiple-noise conditions using deep neural networks. arXiv preprint arXiv:1605.02427 (2016)
  18. 18.
    Liang, S., Srikant, R.: Why deep neural networks for function approximation? arXiv preprint arXiv:1610.04161 (2016)
  19. 19.
    Lu, Z., Pu, H., Wang, F., Hu, Z., Wang, L.: The expressive power of neural networks: a view from the width. In: Advances in Neural Information Processing Systems, pp. 6231–6239 (2017)Google Scholar
  20. 20.
    Ma, X., Zhang, J., Du, B., Ding, C., Sun, L.: Parallel architecture ofconvolutional bi-directional LSTM neural networks for network-wide metroridership prediction. IEEE Trans. Intell. Transp. Syst. 20, 2278–2288 (2018)CrossRefGoogle Scholar
  21. 21.
    Miller, G.F., Todd, P.M., Hegde, S.U.: Designing neural networks using genetic algorithms. In: ICGA, vol. 89, pp. 379–384 (1989)Google Scholar
  22. 22.
    Ray, A., Rajeswar, S., Chaudhury, S.: Text recognition using deep BLSTM networks. In: 2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR), pp. 1–6. IEEE (2015)Google Scholar
  23. 23.
    Seltzer, M.L., Yu, D., Wang, Y.: An investigation of deep neural networks for noise robust speech recognition. In: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 7398–7402. IEEE (2013)Google Scholar
  24. 24.
    Vincent, E., Watanabe, S., Nugraha, A.A., Barker, J., Marxer, R.: An analysis of environment, microphone and data simulation mismatches in robust speech recognition. Comput. Speech Lang. 46, 535–557 (2017)CrossRefGoogle Scholar
  25. 25.
    Vincent, P., Larochelle, H., Lajoie, I., Bengio, Y., Manzagol, P.A.: Stacked denoising autoencoders: learning useful representations in a deep network with a local denoising criterion. J. Mach. Learn. Res. 11(Dec), 3371–3408 (2010)MathSciNetzbMATHGoogle Scholar
  26. 26.
    Weninger, F., Geiger, J., Wöllmer, M., Schuller, B., Rigoll, G.: Feature enhancement by deep LSTM networks for ASR in reverberant multisource environments. Comput. Speech Lang. 28(4), 888–902 (2014)CrossRefGoogle Scholar
  27. 27.
    Weninger, F., Watanabe, S., Tachioka, Y., Schuller, B.: Deep recurrent de-noising auto-encoder and blind de-reverberation for reverberated speech recognition. In: 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 4623–4627. IEEE (2014)Google Scholar
  28. 28.
    Zen, H., Sak, H.: Unidirectional long short-term memory recurrent neural network with recurrent output layer for low-latency speech synthesis. In: 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 4470–4474. IEEE (2015)Google Scholar
  29. 29.
    Zhao, Y., Wang, Z.Q., Wang, D.: Two-stage deep learning for noisy-reverberant speech enhancement. IEEE/ACM Trans. Audio Speech Lang. Process. 27(1), 53–62 (2019)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.PRIS-Lab, Escuela de Ingeniería EléctricaUniversidad de Costa RicaSan PedroCosta Rica

Personalised recommendations