Phonetics and Speech Sciences (말소리와 음성과학)

Korean Society of Speech Sciences (한국음성학회)
권호 표지
DOI QR코드

DOI QR Code

Acoustic Driving Simulator Design for Evaluating an In-car Speech Recognizer

  • Lee, Seongjae (Department of Electrical Engineering, Korea University) ;
  • Kang, Sunmee (Department of Electronic Engineering, Seokyeong University)
  • Received : 2013.01.01
  • Accepted : 2013.03.30
  • Published : 2013.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

This paper is on designing an indoor driving simulator to evaluate the performance of in-car speech recognizer when influenced by the elements, which lower the success rate of speech recognition. The proposed simulator simulates vehicle noise which was pre-recorded in diverse driving environments and driver's speech. Additionally, the proposed Lombard effect conversion module in this simulator enables the speech recorded in a studio environment to convert into various possible driving scenarios. The relevant experimental results have confirmed that the proposed simulator is a feasible approach for realizing an effective method as it achieved similar speech recognition results to the real driving environment.

Keywords

References

  1. Lee, S. J. & Kang S. M. (2012). A study of automatic evaluation platform for speech recognition engine in the vehicle environment. The Journal of Korea Information and Communications Society, 37(7), 538-543. https://doi.org/10.7840/KICS.2012.37.7C.538
  2. Huang, D. Y., Rahardja, S. & Ong, E. P. (2010). Lombard effect Mimicking. ISCA Workshop on Speech Synthesis, 22-24.
  3. Wakao, A. & Takeda, K. & Itakura, F. (1996). Variability of Lombard effects under different noise conditions. Spoken Language, ICSLP Proceedings, 4, 2009-2012.
  4. Hansen, H. L. (2009). Analysis and compensation of Lombard speech across noise type and levels with application to in-set/out-of-set speaker recognition. IEEE Transactions on Audio, Speech, and Language Processing, 17(2), 366-378. https://doi.org/10.1109/TASL.2008.2009019
  5. Bapineedu, G. (2004). Analysis of Lombard effect speech and its application in speaker verification for imposter detection. M.S. Thesis, Hyderabad University.
  6. Marvin, C. & McCallum & John, L. & Campbell (2004). Speech recognition and in-vehicle telematics devices: Potential reductions in driver distraction. International Journal of Speech Technology, 7, 25-33. https://doi.org/10.1023/B:IJST.0000004804.85334.35
  7. Priscilla, L. (2008) The Lombard effect as a communicative phenomenon. UC Berkeley Phonology Lab Annual Report.
  8. Song, H. J. & Jeon, H. B. & Kim, H. S. (2009) Fast speaker adaptation based on Eigenspace-based MLLR using artificially distorted speech in car noise environment. Journal of the Korean Society of Speech Sciences, 4(1), 119-125.