arrow
Volume 8, Issue 1
A Preliminary Study on the Feature Distribution of Deceptive Speech Signals

Xinyu Pan, Heming Zhao, Yan Zhou, Cheng Fan, Wei Zou, Zhiqiang Ren & Xueqin Chen

Journal of Fiber Bioengineering & Informatics, 8 (2015), pp. 179-193.

Published online: 2015-08

Export citation
  • Abstract
A preliminary study is conducted to compare the feature distribution between normal and deceptive speech, and the results are reported in this paper. The objective of this research is to show that deceptive speech may be recognized through the acoustic parameters of general speech characteristics. Six speech parameters, i.e., Mel-frequency Cepstral Coefficients (MFCC), Relative Spectral Filter Perceptual Linear Prediction (RASTA-PLP), pitch frequency, time-domain samples, zero-crossing rate and fractal dimension are used in the statistics. The distributions of these parameters indicate clear differences between the two speech styles. The lowest average degree of difference for these features was 4.74%, and the highest degree was over 20%. Therefore, the distribution demonstrates that there is significant distinction between speech relating the truth and speech relating falsehoods. Linear Discriminant Analysis (LDA) and the Gaussian Mixture Model (GMM) are used to recognize the two psychological states of people's pronunciation, with accuracy above 50%. The results show that there is in fact deceptive information in speech signals and that it can be detected by pattern recognition. These findings provide the theoretical basis for detecting deception in speech signals.
  • AMS Subject Headings

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address
  • BibTex
  • RIS
  • TXT
@Article{JFBI-8-179, author = {}, title = {A Preliminary Study on the Feature Distribution of Deceptive Speech Signals}, journal = {Journal of Fiber Bioengineering and Informatics}, year = {2015}, volume = {8}, number = {1}, pages = {179--193}, abstract = {A preliminary study is conducted to compare the feature distribution between normal and deceptive speech, and the results are reported in this paper. The objective of this research is to show that deceptive speech may be recognized through the acoustic parameters of general speech characteristics. Six speech parameters, i.e., Mel-frequency Cepstral Coefficients (MFCC), Relative Spectral Filter Perceptual Linear Prediction (RASTA-PLP), pitch frequency, time-domain samples, zero-crossing rate and fractal dimension are used in the statistics. The distributions of these parameters indicate clear differences between the two speech styles. The lowest average degree of difference for these features was 4.74%, and the highest degree was over 20%. Therefore, the distribution demonstrates that there is significant distinction between speech relating the truth and speech relating falsehoods. Linear Discriminant Analysis (LDA) and the Gaussian Mixture Model (GMM) are used to recognize the two psychological states of people's pronunciation, with accuracy above 50%. The results show that there is in fact deceptive information in speech signals and that it can be detected by pattern recognition. These findings provide the theoretical basis for detecting deception in speech signals.}, issn = {2617-8699}, doi = {https://doi.org/10.3993/jfbi03201518}, url = {http://global-sci.org/intro/article_detail/jfbi/4698.html} }
TY - JOUR T1 - A Preliminary Study on the Feature Distribution of Deceptive Speech Signals JO - Journal of Fiber Bioengineering and Informatics VL - 1 SP - 179 EP - 193 PY - 2015 DA - 2015/08 SN - 8 DO - http://doi.org/10.3993/jfbi03201518 UR - https://global-sci.org/intro/article_detail/jfbi/4698.html KW - Deceptive Speech KW - Feature Distribution KW - MFCC KW - RASTA-PLP KW - LDA KW - GMM AB - A preliminary study is conducted to compare the feature distribution between normal and deceptive speech, and the results are reported in this paper. The objective of this research is to show that deceptive speech may be recognized through the acoustic parameters of general speech characteristics. Six speech parameters, i.e., Mel-frequency Cepstral Coefficients (MFCC), Relative Spectral Filter Perceptual Linear Prediction (RASTA-PLP), pitch frequency, time-domain samples, zero-crossing rate and fractal dimension are used in the statistics. The distributions of these parameters indicate clear differences between the two speech styles. The lowest average degree of difference for these features was 4.74%, and the highest degree was over 20%. Therefore, the distribution demonstrates that there is significant distinction between speech relating the truth and speech relating falsehoods. Linear Discriminant Analysis (LDA) and the Gaussian Mixture Model (GMM) are used to recognize the two psychological states of people's pronunciation, with accuracy above 50%. The results show that there is in fact deceptive information in speech signals and that it can be detected by pattern recognition. These findings provide the theoretical basis for detecting deception in speech signals.
Xinyu Pan, Heming Zhao, Yan Zhou, Cheng Fan, Wei Zou, Zhiqiang Ren & Xueqin Chen. (2019). A Preliminary Study on the Feature Distribution of Deceptive Speech Signals. Journal of Fiber Bioengineering and Informatics. 8 (1). 179-193. doi:10.3993/jfbi03201518
Copy to clipboard
The citation has been copied to your clipboard