Survey on speech emotion recognition: Features, classification schemes, and databases

Ayadi, Moataz El; Kamel, Mohamed S.; Karray, Fakhri

doi:10.1016/j.patcog.2010.09.020

Cited by 1,719 publications

(859 citation statements)

References 90 publications

Supporting

Mentioning

747

Contrasting

Unclassified

Order By: Relevance

“…Even humans may have difficulty describing how they feel, distinguishing between emotions or remembering how they felt only minutes earlier. A survey verifies the differences among human emotional perception [3], whereas a more detailed study is available in [15]. For the latter case, 20 subjects describe their perception of 6 emotions, namely happiness, hot anger, neutral, interest, panic, and sadness from the Emotional Prosody Speech and Transcripts (EPST) corpus.…”

Section: Related Workmentioning

confidence: 99%

“…The importance of context information is emphasized in [12]. In an additional survey it is stated that besides features, also the applied classifier plays a significant role in emotion recognition performance [3]. For example, Gaussian mixture models (GMMs) cannot model the temporal structure of the data, whereas artificial neural networks classification accuracy seems to be fairly low when compared to other classifiers.…”

Section: Related Workmentioning

confidence: 99%

“…For example, Gaussian mixture models (GMMs) cannot model the temporal structure of the data, whereas artificial neural networks classification accuracy seems to be fairly low when compared to other classifiers. On the contrary, SVMs appear to have global optimality of the training algorithm as well as highperformance data-dependent generalization bounds [3].…”

Section: Related Workmentioning

confidence: 99%

“…In other words, it is not possible for the classifier to be tested on utterances derived from the same speaker whose utterances belong to the training set. In [3], an alternative method for speaker-independency is proposed. In specific the authors mention that a speaker-independent emotion recognition system could be implemented as a combination of a speaker identification system followed by a speaker-dependent emotion recognition system.…”

Section: Related Workmentioning

confidence: 99%

See 3 more Smart Citations

Speaker-independent emotion recognition exploiting a psychologically-inspired binary cascade classification schema

Kotti

Paternò

2012

Int J Speech Technol

View full text Add to dashboard Cite

In this paper, a psychologically-inspired binary cascade classification schema is proposed for speech emotion recognition. Performance is enhanced because commonly confused pairs of emotions are distinguishable from one another. Extracted features are related to statistics of pitch, formants, and energy contours, as well as spectrum, cepstrum, perceptual and temporal features, autocorrelation, MPEG-7 descriptors, Fujisaki's model parameters, voice quality, jitter, and shimmer. Selected features are fed as input to K nearest neighborhood classifier and to support vector machines. Two kernels are tested for the latter: linear and Gaussian radial basis function. The recently proposed speaker-independent experimental protocol is tested on the Berlin emotional speech database for each gender separately. The best emotion recognition accuracy, achieved by support vector machines with linear kernel, equals 87.7%, outperforming state-of-the-art approaches. Statistical analysis is first carried out with respect to the classifiers' error rates and then to evaluate the information expressed by the classifiers' confusion matrices.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Speaker-independent emotion recognition exploiting a psychologically-inspired binary cascade classification schema

Kotti

Paternò

2012

Int J Speech Technol

View full text Add to dashboard Cite

show abstract

“…So to make these two binary images mutually exclusive we define the following (6) Here is a weak edge response and is a strong edge response So if we combine wear edge response to the strong edge response then we will get valid edge points so final image result is…”

Section: Aslc Edge Detection Marr-hildreth Hystresis Analysis Edge Fementioning

confidence: 99%