On noise and feature settings action on quality of the automatic speech recognition system

Article Sidebar

PDF (Русский)
Published: May 15, 2013
DOI: https://doi.org/10.20535/2312-1807.2013.18.1.189185
Keywords:
speech recognition, noise, noise reduction, parameterization, transcription, speech corpus.

Main Article Content

S. A. Yolkin
National technical university of Ukraine "Igor Sikorsky Kyiv polytechnic institute"
A. G. Liashenko
V. P. Ovsyanik
A. N. Prodeus

Abstract

The development of automatic speech recognition systems is expedient to make with the different nature of interference (noise acoustic environment, reverberation, filtering and encoding of speech in communication systems). The effect of noise and some features of the automatic speech recognition system on the recognition quality is experimentally investigated in this paper. Recommendations on the parameters optimization of the automatic speech recognition system for several scenarios of usage are obtained.

Reference 10, figures 1, tables 6.

Article Details

How to Cite
Yolkin, S. A., Liashenko, A. G., Ovsyanik, V. P., & Prodeus, A. N. (2013). On noise and feature settings action on quality of the automatic speech recognition system. Electronics and Communications, 18(1), 94–101. https://doi.org/10.20535/2312-1807.2013.18.1.189185
Issue
Vol. 18 No. 1 (2013)
Section
Acoustical devices and systems
Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Authors who publish with this journal agree to the following terms:

  1. Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
  2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
  3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).

References

Audacity / [Online]. Available at: http://audacity.sourceforge.net. (Rus) (17.01.2013)

Documentation for HTK [Online]. Available at: http://htk.eng.cam.ac.uk/docs/docs.shtml. (17.01.2013).

Furui S. 50 years of progress in speech and speaker recognition [Online]. – Available at: http://www.furui.cs.titech.ac.jp/publication/2005/SPCOM05.pdf. (17.01.2013)

Nelson N. HTK MFCC Study. [Online]. Available at: http://speech-research.com/mfccStudy.html (17.01.2013).

Sony Sound Forge™ 10 Pro [Online]. Available at: http://www.sonycreativesoftware.com/.(17.01.2013)

Verbout S.M. Signal (1994), [Enhancement for Automatic Recognition of Noisy Speech]. Master degree dissertation for the degree of Master of Science. May, P.84.

A short course on HTK [Online]. Available at: http://www.speech.com.ua/htk_course.html. (Rus) (17.01.2013).

Prodeus А.N., Ladoshko O.N. (2007), [Optimization of speech recognition system algorithms with usage of НТК toolkit]. Electronics and Communication. no 4. pp. 53–60. (Rus)

Prodeus А.N., Litvinov S.V. (2009), [Modeling of Ukrainian speech recognition system with usage of НТК toolkit]. Electronics and Communication. no 1. pp. 88–94. (Rus)

Fonetic transcription. Modern Russian language [Online]. Available at: http://morfema.ru/publ/15-1-0-3. (Rus) (17.01.2013).