13th Annual Conference of the International Speech Communication Association

Portland, OR, USA
September 9-13, 2012

Multi-System Fusion of Extended Context Prosodic and Cepstral Features for Paralinguistic Speaker Trait Classification

Michelle Hewlett Sanchez, Aaron Lawson, Dimitra Vergyri, Harry Bratt

Speech Technology and Research Laboratory, SRI International, Menlo Park, CA, USA

As automatic speech processing has matured, research attention has expanded to paralinguistic speech problems that aim to detect "beyondthe- words" information. This paper focuses on the identification of seven speaker trait categories from the Interspeech Speaker Trait Challenge: likeability, intelligibility, openness, conscientiousness, extraversion, agreeableness, and neuroticism. Our approach combines multiple features including prosodic, cepstral, shifted-delta cepstral, and a reduced set of the OpenSMILE features. Our classification approaches included GMMUBM, eigenchannel, support vector machines, and distance based classifiers. Optimized feature reduction and logistic regression-based score calibration and fusion led to results that perform competitively against the challenge baseline in all categories.

Index Terms: speaker traits, prosody, MFCCs, Gaussian mixture modeling

Full Paper

Bibliographic reference.  Sanchez, Michelle Hewlett / Lawson, Aaron / Vergyri, Dimitra / Bratt, Harry (2012): "Multi-system fusion of extended context prosodic and cepstral features for paralinguistic speaker trait classification", In INTERSPEECH-2012, 514-517.