Paper Info
Title
Voice analysis for detection of deception
Abstract
Human voice can signal the psychological state of an individual. Be it emotional, physical, intentional or unconscious, voice reveals the internal state of the speaker. Therefore, the main aim of the present study is to identify which voice parameter could be used as a reliable indicator of stress due to deception. For analyzing real life natural stress, the voice of crime suspects was recorded during ongoing interrogations at the police station. PRAAT software was used for analysis. Results indicated a significant increase in the mean fundamental frequency (F <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> ) and formant frequencies F1, F2. Marginal decrease in jitter was observed, however, shimmer did not at all correlate to stress due to deception. Pitch contours indicated considerable rise in the pitch of crime suspects while answering specific direct questions. Thus, it was concluded that mean F <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> and formant frequencies F1, F2 correlate strongly to stress.
Year
DOI
Venue
2016
10.1109/KICSS.2016.7951455
2016 11th International Conference on Knowledge, Information and Creativity Support Systems (KICSS)
Keywords
Field
DocType
Voice stress analysis,detection of deception,fundamental frequency,jitter,shimmer,spectrogram,pitch contour
Voice analysis,Fundamental frequency,Human voice,Computer science,Deception,Spectrogram,Speech recognition,Jitter,Formant
Conference
ISBN
Citations 
PageRank 
978-1-5090-5131-1
0
0.34
References 
Authors
2
4
Name
Order
Citations
PageRank
Savita Sondhi100.34
Ritu Vijay2165.90
Munna Khan3185.12
Ashok Kumar Salhan421.86