Abstract
A comparative study of speech recognition performance among systems trained with manually labeled corpora and systems trained with semiautomatically labeled corpora is introduced. An automatic labeling system was designed to generate phoneme labels files for all words within the corpus used to train a system of automatic speech recognition. Speech recognition experiments were performed using the same corpus, first training with manually, and later with automatically generated labels. Results show that the recognition performance is better when the training of selected diccionary, is made with automatic label files than when it is made with manual label files. Not only is the automatic labeling of speech corpora faster than manual labeling, but also it is free from the subjectivity inherent in the manual segmentation performed by specialists. The performance achieved in this work is greater than 96 %.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Becchetti, C., Ricotti, L.C.: Speech Recognition. Theory and C ++ Implementation, pp. 305–307. Wiley, New York (1999)
Fantinato, P.C., et al.: A Fractal-based approach for speech segmentation. In: Tenth IEEE International Symposium on Multimedia, ISM 2008, 15–17 December 2008, pp. 551–555 (2008). doi:10.1109/ISM.2008.123
Galka, J., Ziolko, M.: Wavelets in speech segmentation. In: The 14th IEEE Mediterranean Electrotechnical Conference, MELECON 2008, 5–7 May 2008, pp. 876–879 (2008)
Hieronymus, J.L.: ASCII Phonetic Symbols for the World’s Languages: Worldbet, pp. 43–44 (1993). http://www.ling.ohio-state.edu/~edwards/WorldBet/worldbet.pdf
Hosom, J.P.: Automatic time alignment of phonemes using acoustic-phonetic information. Oregon Graduate Institute of Science and Technology. Ph.D. thesis, pp. 121–122 (2000)
Lander, T.: The CSLU Labeling Guide. Center of Spoken Language Understanding, Oregon Graduate Institute, pp. 52–56 (1997). http://www.cslu.ogi.edu/corpora/docs/labeling.pdf
Pineda, L.A., et al.: The Corpus DIMEx100: transcription and evaluation. Lang. Resour. Eval. 44, 347–370 (2009, 2010). doi:10.1007/s10579-009-9109-9
Bansal, P., et al.: Speech synthesis – automatic segmentation. Int. J. Comput. Appl. (0975–8887) 98 (4), (2014)
Toledano, D.T., Gómez, L.A.H., Grande, L.V.: Automatic phonetic segmentation. IEEE Trans. Speech Audio Process. 11(6), 617–625 (2003). doi:10.1109/TSA.2003.813579
Ziolko, B., Manandhar, S., Wilson, R.C.: Phoneme segmentation of speech. In: 18th International Conference on Pattern Recognition, ICPR 2006, vol. 4, pp. 282–285 (2006). doi:10.1109/ICPR.2006.931
Acknowledgments
We thank National Polytechnic Institute (IPN - Instituto Politécnico Nacional, México), COFAA-IPN, PIFI-IPN SIP-IPN 20141454 and SIP-IPN 20130617; for their academic and financial support.
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
Sergio, SG., Cristian-Remington, JM., Luis, OR.J. (2015). Automatic Phoneme Border Detection to Improve Speech Recognition. In: Sidorov, G., Galicia-Haro, S. (eds) Advances in Artificial Intelligence and Soft Computing. MICAI 2015. Lecture Notes in Computer Science(), vol 9413. Springer, Cham. https://doi.org/10.1007/978-3-319-27060-9_11
Download citation
DOI: https://doi.org/10.1007/978-3-319-27060-9_11
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-27059-3
Online ISBN: 978-3-319-27060-9
eBook Packages: Computer ScienceComputer Science (R0)