A Neural Network Model for Phoneme Generation

Abstract:

The paper discusses the possibility of phonemes generation based on a recurrent neural network model. In each phoneme a typical or elemental pattern can be identified that repeats itself with slight fluctuations along the signal length. This elemental pattern constitutes the training data for the recurrent neural network. After training, the network can generate three new periods of elemental patterns. In a repetitive loop the network can generate the entire phoneme signal. The model proved very simple and effective, and the generated phonemes gave the impression of a natural sound.

• Periodical: Applied Mechanics and Materials (Volume 367)
• Pages: 478-483
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.367.478
• Online since: August 2013
• Authors: Marius Crisan
• Keywords: Neural Network Models, Phoneme Generation, Speech Processing, Time Series Modeling

References

[1] J. Holmes and W. Holmes: Speech Synthesis and Recognition, 2nd Edition, Taylor & Francis, N.Y. (2001).

[2] D. Jurafsky and J.H. Martin: Speech and Language Processing. An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition, Second Edition, Pearson Prentice Hall, (2008).

DOI: 10.1162/coli.b09-001

[3] S. McLaughlin and P. Maragos: Nonlinear methods for speech analysis and synthesis. In: Marshall S, Sicuranza G, editor. Advances in nonlinear signal and image processing, Vol. 6. Hindawi Publishing Corporation (2007), p.103.

[4] V. Pitsikalis and P. Maragos: Analysis and classification of speech signals by generalized fractal dimension features, Speech Communication, Vol. 51, no. 12, (2009), p.1206–1223.

DOI: 10.1016/j.specom.2009.06.005

[5] K. Sreenivasa Rao: Role of neural network models for developing speech systems, Sadhana Vol. 36, Part 5, (2011 Oct), p.783–836.

DOI: 10.1007/s12046-011-0047-z

[6] E.V. Raghavendra, P. Vijayaditya and K. Prahallad: Speech synthesis using artificial neural networks, National Conference on Communications (NCC), Chennai, India, (2010), pp.1-5.

DOI: 10.1109/ncc.2010.5430190

[7] O. Karaali, G. Corrigan, and I. Gerson: Speech Synthesis with Neural Networks, World Congress on Neural Networks, San Diego, (1996 Sep), pp.45-50.

[8] W. K. Lo and P. C. Ching: Phone-Based Speech Synthesis With Neural Network And Articulatory Control, In Proceedings of Fourth International Conference on Spoken Language (ICSLP 96), (1996), vol. 4, pp.2227-2230.

DOI: 10.1109/icslp.1996.607248

[9] I. Gerson, O. Karaali and G. Corrigan: Neural Network Speech Synthesis, In Proceedings of the World Conference on Neural Networks, San Diego, California, USA, (1996), pp.45-50.

[10] M. Malcangi and D. Frontini: A Language-Independent Neural Network-Based Speech Synthesizer. In: Neurocomputing, 73: 1-3 (2009 Dec), pp.87-96.

DOI: 10.1016/j.neucom.2008.08.023

[11] S.D. Balkin: Using Recurrent Neural Networks for Time Series Forecasting, Technical Report 97-11, Pennsylvania State University, (1997).

[12] R.J. Frank, N. Davey and S.P. Hunt: Time Series Prediction and Neural Networks. In: Journal of Intelligent and Robotic Systems 31, (2001), pp.91-103.

[13] W. Kinzel: Predicting and generating time series by neural networks: An investigation using statistical physics. In: Computational Statistical Physics (2002), pp.97-111.

DOI: 10.1007/978-3-662-04804-7_6

[14] A. Priel and I. Kanter: Time series generation by recurrent neural networks, In: Annals of Mathematics and Artificial Intelligence 39 (2003), p.315–332.

DOI: 10.1023/a:1024620813258

[15] M. Crisan: New Aspects of Phoneme Synthesis Based on Chaotic Modeling, In: Proceedings of 2011 International Conference on Instrumentation, Measurement, Circuits and Systems (ICIMCS 2011), Hong Kong, pp.605-614.

DOI: 10.1007/978-3-642-27334-6_71