Odyssey 2012 - The Speaker and Language Recognition Workshop

Singapore
June 25-28, 2012

Voice Source Features for Forensic Voice Comparison - an Evaluation of the GLOTTEX Software Package

Ewald Enzinger (1,2), Cuiling Zhang (1,3), Geoffrey Stewart Morrison (1)

(1) Forensic Voice Comparison Laboratory, School of Electrical Engineering & Telecommunications, University of New South Wales, Australia
(2) Acoustics Research Institute, Austrian Academy of Sciences, Austria
(3) Department of Forensic Science & Technology, China Criminal Police University, Shenyang, China

GLOTTEX is a software package which extracts information about voice source properties, including estimates of properties related to physical structures of the vocal folds. It has been proposed that the output of GLOTTEX can be used as part of a forensic-voice-comparison system. We test this using manually labeled segments from a database of voice recordings of 60 female Chinese speakers. Performance was assessed relative to a baseline MFCC GMM-UBM system. GMM-UBM systems based on features extracted by GLOTTEX were combined with the baseline system using logistic-regression fusion. System performance was assessed in three channel conditions: high-quality vs. high-quality, mobile-to-landline vs. mobile-to-landline, and mobile-to-landline vs. high-quality. Substantial improvements over the baseline system were not observed.

Full Paper

Bibliographic reference.  Enzinger, Ewald / Zhang, Cuiling / Morrison, Geoffrey Stewart (2012): "Voice source features for forensic voice comparison - an evaluation of the GLOTTEX software package", In Odyssey-2012, 78-85.