Primitives-based evaluation and estimation of emotions in speech

Grimm, Michael; Kroschel, Kristian; Mower, Emily; Narayanan, Shrikanth

doi:10.1016/j.specom.2007.01.010

Cited by 259 publications

(172 citation statements)

References 26 publications

(30 reference statements)

Supporting

Mentioning

148

Contrasting

Unclassified

Order By: Relevance

“…This result is consistent with some previous studies, which have suggested that emotional dimensions are more suitable to distinguish and describe the vocal expression of emotions than labels of discrete emotions CORNELIUS, 2003;LUGGER;YANG, 2007;PEREIRA, 2000;BARBOSA, 2009). Research on the automatic recognition of emotions from speech and speech synthesis can benefit from this finding, since the use of emotional dimensions may allow the reliable identification and synthesis of more subtle expressions of emotions and changes in speech expressiveness over time (GRIMM, et al, 2007;SCHRÖDER et al, 2001;BARBOSA, 2009).…”

Section: Discussionmentioning

confidence: 99%

“…fundamental frequency and intensity) and this may cause confusion when trying to describe these emotions by means of discrete labels (LUGGER;YANG, 2007;PEREIRA, 2000). Emotional dimensions may be useful for research on speech synthesis and automatic recognition of emotions from speech, since their continuous quality allows the modeling of weaker emotional expressions, as well as gradual changes in speech expressiveness over time (GRIMM et al, 2007;SCHRÖDER et al, 2001;BARBOSA, 2009). …”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Perception of emotional prosody: investigating the relation between the discrete and dimensional approaches to emotions

Silva

Barbosa

2017

RELIN

View full text Add to dashboard Cite

Emotional phenomena can be described according to various psychological approaches, the most adopted being the discrete (basic) and the dimensional ones. This study aimed at investigating the relation between the perception of some basic emotions and emotional dimensions in speech, as well as determining which acoustic cues are related to their perception. We conducted two perception experiments with utterances selected from a foreign language (Swedish) of which the listeners had no knowledge. In the first one, Brazilian subjects rated on 5-point scales the expressivity of four basic emotions: joy, anger, sadness, and calmness. In the second, a distinct group of Brazilian subjects rated the expressivity of five emotional dimensions: activation, fairness, valence, motivation, and involvement. The perception of the basic emotions and of the emotional dimensions was then compared by means of the Spearman's correlation coefficient. The five emotional dimensions were significantly correlated to some extent with the basic emotions, and these correlations were, in general, consistent with the literature and with the hypotheses that guided this study. We also performed an acoustic analysis, Linguagem, Belo Horizonte, v.25, n.3, p. 1075-1102, 2017 1076 in which twelve acoustic parameters were automatically computed for the utterances evaluated by the listeners. The parameters which correlated better with the listeners' judgments were fundamental frequency (median, interquantile semi-amplitude, 99.5% quantile), spectral tilt (mean and standard deviation), and LTAS slope. We concluded that it is possible to describe the perception of basic emotions in speech as a combination of emotional dimensions and that emotional dimensions may be better for describing the expression of emotions in speech. Keywords: emotional prosody; basic emotions; emotional dimensions; perception test. Revista de Estudos da Resumo:Os fenômenos emocionais podem ser descritos de acordo com várias abordagens psicológicas, sendo a discreta (básica) e a dimensional as mais adotadas. Este estudo teve como objetivo investigar a relação entre a percepção de algumas emoções básicas e dimensões emocionais na fala, bem como determinar quais pistas acústicas estão relacionadas com sua percepção. Conduziram-se dois experimentos de percepção com enunciados selecionados de uma língua estrangeira (sueco) da qual os ouvintes não possuíam nenhum conhecimento. No primeiro, sujeitos brasileiros julgaram em escalas de cinco pontos a expressividade de quatro emoções básicas: alegria, raiva, tristeza e calma. No segundo, um grupo distinto de sujeitos brasileiros avaliou a expressividade de cinco dimensões emocionais: ativação, justiça, valência, motivação e envolvimento. A percepção das emoções básicas e das dimensões emocionais foi então comparada por meio do coeficiente de correlação de Spearman. As cinco dimensões emocionais correlacionaram-se significativamente em algum grau com as emoções básicas e essas correlações foram, no geral, consistentes com a lite...

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Perception of emotional prosody: investigating the relation between the discrete and dimensional approaches to emotions

Silva

Barbosa

2017

RELIN

View full text Add to dashboard Cite

show abstract

“…Different researchers proposes values from k = 1 [23], [30] to k = 20 [14]. Usually the value of k is found by trial and error, different values are taken and the performance of these classifiers is compared [16], [17].…”

Section: A K-nnmentioning

confidence: 99%

Benchmarking classification models for emotion recognition in natural speech: A multi-corporal study

Тарасов¹,

Delany²

2011

Face and Gesture 2011

View full text Add to dashboard Cite

Abstract-A significant amount of the research on automatic emotion recognition from speech focuses on acted speech that is produced by professional actors. This approach often leads to overoptimistic results as the recognition of emotion in real-life conditions is more challenging due the propensity of mixed and less intense emotions in natural speech. The paper presents an empirical study of the most widely used classifiers in the domain of emotion recognition from speech, across multiple non-acted emotional speech corpora. The results indicate that Support Vector Machines have the best performance and that they along with Multi-Layer Perceptron networks and k-nearest neighbour classifiers perform significantly better (using the appropriate statistical tests) than decision trees, Naïve Bayes classifiers and Radial Basis Function networks.

show abstract

“…Some studies (see for a review Murray & Arnott (1993) & Scherer (2003) have empirically investigated acoustic and prosodic characteristics such as pitch variables and speaking rate, which are taken into emotion recognition models and this approach has shown varying detection rates (Devillers et al, 2005). Table 1 shows some of the recognition accuracy of empirical evidence in the range of between 50% (Nakatsu et al, 2000) and 83.5% (Grimm et al, 2007). It should be noted that the accuracy may be dependent on the number of emotional states attempted in the studies.…”

Section: Related Literature On Emotion Recognitionmentioning

confidence: 99%

“…Recognition Rates Vocal 50% (Nakatsu et al, 2000),73% (Lee et al, 2006), 83.5% (Grimm et al, 2007) Facial 98% (Essa & Pentland, 1997), 86% (Anderson & McOwan, 2006), 78% (Ioannou et al, 2005) Body gestures 84-92% (Kapur et al, 2005),44-90% (Castellano et al, 2007) Of the above mentioned approaches, a physiological approach is promising in that inner bodily changes are reflected in autonomous nervous systems and thus, integrally related to human emotion (Picard et al, 2001;Lisetti et al, 2003;Zhihong et al, 2007). For example, there is empirical evidence that physiological activities in face, finger, and body (e.g., EMG, PPG, EEG) are related to emotions.…”

Section: Approachesmentioning

confidence: 99%