Two channels have been distinguished in human interaction: one transmits explicit messages, which may be about anything or nothing; the other transmits implicit messages about the speakers themselves. Both linguistics and technology have invested enormous efforts in understanding the first, explicit channel, but the second is not as well understood. Understanding the other party's emotions is one of the key tasks associated with the second, implicit channel. To tackle that task, signal processing and analysis techniques have to be developed, while, at the same time, consolidating psychological and linguistic analyses of emotion. This article examines basic issues in those areas. It is motivated by the PKYSTA project, in which we aim to develop a hybrid system capable of using information from faces and voices to recognize people's emotions.

Emotion recognition in human-computer interaction

Professional actors' portrayals of 14 emotions varying in intensity and valence were presented to judges. The results on decoding replicate earlier findings on the ability of judges to infer vocally expressed emotions with much-better-than-chance accuracy, including consistently found differences in the recognizability of different emotions. A total of 224 portrayals were subjected to digital acoustic analysis to obtain profiles of vocal parameters for different emotions. The data suggest that vocal parameters not only index the degree of intensity typical for different emotions but also differentiate valence or quality aspects. The data are also used to test theoretical predictions on vocal patterning based on the component process model of emotion (K.R. Scherer, 1986). Although most hypotheses are supported, some need to be revised on the basis of the empirical evidence. Discriminant analysis and jackknifing show remarkably high hit rates and patterns of confusion that closely mirror those found for listener-judges.

/pdf/acoustic-profiles-in-vocal-emotion-expression-4whpq8okig.pdf

Acoustic profiles in vocal emotion expression.

Work on voice sciences over recent decades has led to a proliferation of acoustic parameters that are used quite selectively and are not always extracted in a similar fashion. With many independent teams working in different research areas, shared standards become an essential safeguard to ensure compliance with state-of-the-art methods allowing appropriate comparison of results across studies and potential integration and combination of extraction and recognition systems. In this paper we propose a basic standard acoustic parameter set for various areas of automatic voice analysis, such as paralinguistic or clinical speech analysis. In contrast to a large brute-force parameter set, we present a minimalistic set of voice parameters here. These were selected based on a) their potential to index affective physiological changes in voice production, b) their proven value in former studies as well as their automatic extractability, and c) their theoretical significance. The set is intended to provide a common baseline for evaluation of future research and eliminate differences caused by varying parameter sets or even different implementations of the same parameters. Our implementation is publicly available with the openSMILE toolkit. Comparative evaluations of the proposed feature set and large baseline feature sets of INTERSPEECH challenges show a high performance of the proposed set in relation to its size.

/pdf/the-geneva-minimalistic-acoustic-parameter-set-gemaps-for-36spb4mksx.pdf

The Geneva Minimalistic Acoustic Parameter Set (GeMAPS) for Voice Research and Affective Computing

Over 3 million teachers in the United States use their voice as a primary tool of trade and are thought to be at higher risk for occupation-related voice disorders than the general population. However, estimates regarding the prevalence of voice disorders in teachers and the general population vary considerably. To determine the extent that teachers are at greater risk for voice disorders, 2,531 randomly selected participants from Iowa and Utah (1,243 teachers and 1,288 nonteachers) were interviewed by telephone using a voice disorder questionnaire. Prevalence-the number of cases per population at risk at a specific time-was determined. The prevalence of reporting a current voice problem was significantly greater in teachers compared with nonteachers (11.0% vs. 6.2%), chi(2)(1) = 18.2, p 4 weeks in duration), compared with acute voice disorders (20.9% vs. 13.3%), chi(2)(1) = 8.7, p =.003. To assess the association between past voice disorders and possible risks, adjusted odds ratios (ORs) were estimated using multiple logistic regression. The results identified that being a teacher, being a woman, being between 40 and 59 years of age, having 16 or more years of education, and having a family history of voice disorders were each positively associated with having experienced a voice disorder in the past. These results support the notion that teaching is a high-risk occupation for voice disorders. Important information is also provided regarding additional factors that might contribute to the development of voice disorders.

Prevalence of Voice Disorders in Teachers and the General Population

The reliability of listeners’ ratings of voice quality is a central issue in voice research because of the clinical primacy of such ratings and because they are the standard against which other mea...

Perceptual Evaluation of Voice Quality

A set of 17 voices representing various voice disorders were presented to a jury of voice experts who rated them on a 5-point scale for each of 28 terms frequently used for describing voices. A statistical treatment of these ratings yielded 5 bipolar factors accounting for 85% of the total variance. Significant correlations were found between four of these factors and certain acoustic data extracted fron long-time-average spectra and fundamental frequency distribution analysis of the same voice sample.

Perceptual and acoustic correlates of abnormal voice qualities.

Information on diagnosis, occupation, sex and age of new voice patients seen during a period of six months during 1992–93 were reported from the eight hospital departments of phoniatrics in Sweden. Data from altogether 1212 patients of working age were obtained. Phonasthenia was by far the most common diagnosis. Teaching professions were more common than any other occupational group. There were twice as many women as men seeking phoniatric care for voice problems. When a comparison was made with data from the Swedish Population and Housing Census 1990, it turned out that among the voice patients the teaching professions were clearly overrepresented, and so were social workers, lawyers and clergymen. The need for improved preventive voice care in the training programmes for these professions is quite clear.

Björn Fritzell

Papers

Perceptual and acoustic correlates of abnormal voice qualities.

Voice disorders and occupations

Experiences with recurrent laryngeal nerve section for spastic dysphonia.

Long-term results of recurrent laryngeal nerve resection for adductor spasmodic dysphonia

International association of logopedics and phoniatrics (IALP) voice committee discussion of assessment topics