Hierarchical Clustering and Classification of Emotions in Human Speech Using Confusion Matrices

Reyes-Vargas, Manuel; Sánchez-Gutiérrez, Máximo Eduardo; Rufiner, Hugo Leonardo; Albornoz, Enrique Marcelo; Vignolo, Leandro D.; Licona, Fabiola Martínez; Goddard-Close, John

doi:10.1007/978-3-319-01931-4_22

Cited by 4 publications

(3 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nevertheless, similarity between neural and experiential data would suggest similar representational segregation in the neural code and subjective feelings. From the group-averaged confusion matrix, we calculated a distance matrix by taking the category confusion vectors for each pair of emotions and by calculating the Euclidean distance between these vectors (see Reyes-Vargas et al , 2013 ). We then employed hierarchical cluster analysis in Matlab to investigate how different emotions cluster together based on their neural similarities.…”

Section: Methodsmentioning

confidence: 99%

Distributed affective space represents multiple emotion categories across the human brain

Saarimäki

Ejtehadian

Glerean

et al. 2018

Social Cognitive and Affective Neuroscience

118

View full text Add to dashboard Cite

The functional organization of human emotion systems as well as their neuroanatomical basis and segregation in the brain remains unresolved. Here, we used pattern classification and hierarchical clustering to characterize the organization of a wide array of emotion categories in the human brain. We induced 14 emotions (6 ‘basic’, e.g. fear and anger; and 8 ‘non-basic’, e.g. shame and gratitude) and a neutral state using guided mental imagery while participants' brain activity was measured with functional magnetic resonance imaging (fMRI). Twelve out of 14 emotions could be reliably classified from the haemodynamic signals. All emotions engaged a multitude of brain areas, primarily in midline cortices including anterior and posterior cingulate gyri and precuneus, in subcortical regions, and in motor regions including cerebellum and premotor cortex. Similarity of subjective emotional experiences was associated with similarity of the corresponding neural activation patterns. We conclude that different basic and non-basic emotions have distinguishable neural bases characterized by specific, distributed activation patterns in widespread cortical and subcortical circuits. Regionally differentiated engagement of these circuits defines the unique neural activity pattern and the corresponding subjective feeling associated with each emotion.

show abstract

Section: Methodsmentioning

confidence: 99%

Distributed affective space represents multiple emotion categories across the human brain

Saarimäki

Ejtehadian

Glerean

et al. 2018

Social Cognitive and Affective Neuroscience

118

View full text Add to dashboard Cite

show abstract

“…The MFCCs, despite the limitations of the Mel filterbank, are the most widely used features for most speech processing applications like speech recognition ( Deller et al, 1993 ), speaker verification ( Sahidullah and Saha, 2013 ), emotion recognition ( Ooi et al, 2014;Zheng et al, 2014;Reyes-Vargas et al, 2013;Ververidis and Kotropoulos, 2006 ), language recognition ( Huang et al, 2013 ), etc., and even for non-speech acoustic signal processing tasks, such as music information retrieval ( Qin et al, 2013 ). However, as discussed in the preceding subsection, it is quite ambitious to assume that they would provide the best possible performance for all applications.…”

Section: Utility Of Non-conventional Analysis and Filterbank Optimiza...mentioning

confidence: 99%

Empirical Mode Decomposition for adaptive AM-FM analysis of Speech: A Review

Sharma

Vignolo

Schlotthauer

et al. 2017

Speech Communication

View full text Add to dashboard Cite

“…We next investigated the similarities between emotions using the category confusions from the whole-brain classification. From the group-averaged confusion matrix, we calculated a distance matrix by taking the category confusion vectors for each pair of emotions and by calculating the Euclidean distance between these vectors (see Reyes-Vargas et al, 2013). We then employed hierarchical cluster analysis in Matlab to investigate how different emotions cluster together based on their neural similarities.…”

Section: Hierarchical Clusteringmentioning

confidence: 99%

Distributed affective space represents multiple emotion categories across the brain

Saarimäki

Ejtehadian

Glerean

et al. 2017

Preprint

View full text Add to dashboard Cite

The functional organization of human emotion systems as well as their neuroanatomical basis and segregation in the brain remains unresolved. Here we used pattern classification and hierarchical clustering to reveal and characterize the organization of discrete emotion categories in the human brain. We induced 14 emotions (6 "basic", such as fear and anger; and 8 "non-basic", such as shame and gratitude) and a neutral state in participants using guided mental imagery while their brain activity was measured with functional magnetic resonance imaging (fMRI). Twelve out of 14 emotions could be reliably classified from the fMRI signals. All emotions engaged a multitude of brain areas, primarily in midline cortices including anterior and posterior cingulate and precuneus, in subcortical regions, and in motor regions including cerebellum and premotor cortex. Similarity of subjective emotional experiences was associated with similarity of the corresponding neural activation patterns. We conclude that the emotions included in the study have discrete neural bases characterized by specific, distributed activation patterns in widespread cortical and subcortical circuits, and highlight both overlaps and differences in the locations of these for each emotion. Locally differentiated engagement of these globally shared circuits defines the unique neural fingerprint activity pattern and the corresponding subjective feeling associated with each emotion.

show abstract

Hierarchical Clustering and Classification of Emotions in Human Speech Using Confusion Matrices

Cited by 4 publications

References 17 publications

Distributed affective space represents multiple emotion categories across the human brain

Distributed affective space represents multiple emotion categories across the human brain

Empirical Mode Decomposition for adaptive AM-FM analysis of Speech: A Review

Distributed affective space represents multiple emotion categories across the brain

Contact Info

Product

Resources

About