Speaker state recognition using an HMM-based feature extraction method

Gajsek, Rok; Mihelic, F.; Dobrišek, Simon

doi:10.1016/j.csl.2012.01.007

Cited by 25 publications

(10 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A work on Speaker state recognition using an HMM-based feature extraction method [4] done by R. Gajˇsek * , F. Miheliˇc, S. Dobriˇsek. This system uses acoustic features for recognizing various paralinguistic phenomena.…”

Section: Related Workmentioning

confidence: 99%

Hidden MARKOV Model Approach Towards Emotion Detection From Speech Signal

Sathiyamurthy¹,

Pavidhra²,

Monisha³

et al. 2015

Computer Science &Amp; Information Technology ( CS &Amp; IT )

View full text Add to dashboard Cite

Emotions carry the token indicating a human's mental state. Understanding the emotion exhibited becomes difficult for people suffering from autism and alexithymia. Assessment of emotions can also be beneficial in interactions involving a human and a machine. A system is developed to recognize the universally accepted emotions such as happy, anger, sad, disgust, fear and surprise. The gender of the speaker helps to obtain better clarity for identifying the emotion. Hidden Markov Model serves the purpose of gender identification.

show abstract

Section: Related Workmentioning

confidence: 99%

Hidden MARKOV Model Approach Towards Emotion Detection From Speech Signal

Sathiyamurthy¹,

Pavidhra²,

Monisha³

et al. 2015

Computer Science &Amp; Information Technology ( CS &Amp; IT )

View full text Add to dashboard Cite

show abstract

“…Still, most of the aforementioned phoneme-level modeling emotion classification techniques used forced alignment or manual annotation for the extraction of the phoneme borders. Just some methods faced real-life conditions by using ASR engines for generating the phoneme alignment [16]. Current ASR techniques are not able to provide as good phoneme alignment on affective speech samples as manual annotation or forced alignment.…”

Section: Introductionmentioning

confidence: 99%

“…According to Batliner et al [10] words can be seen as the smallest possible chunk for analysis. A comparably smaller number of classification techniques are based on phonetic pattern modeling within emotion classification [11], [12], [13], [14], [15], [16], [17]. Still, most of the aforementioned phoneme-level modeling emotion classification techniques used forced alignment or manual annotation for the extraction of the phoneme borders.…”

Section: Introductionmentioning

confidence: 99%

Determining the Smallest Emotional Unit for Level of Arousal Classification

Vlasenko

Wendemuth

2013

2013 Humaine Association Conference on Affective Computing and Intelligent Interaction

View full text Add to dashboard Cite

Most state-of-the-art emotion recognition methods are based on turn-and frame-level analysis independent from phonetic transcription. Currently "affective computing" community could not specify the smallest emotional standard unit which can be easily classified and determined by any "advanced" and "non-advanced" listener. It is known that, acoustic modeling on the smallest phonetic unit (phoneme) started a new era in automatic speech recognition: switch from speaker dependent isolated word recognition to speaker independent continuous speech recognition. In or current research we showed that phoneme can be used as as smallest unit for high and low arousal emotion classification task. We trained our classifications models on the VAM dataset material and evaluated them on speech samples from the DES dataset. For our experiments we employed two different emotion classification approaches: general (phonetic pattern independent) and phoneme-based (phonetic pattern dependent). Both classification approaches used MFFC features extracted on the frame level. Our experimental results impressively show that the proposed phoneme-based classification technique could increase emotion classification performance by about 9.68% absolute (15.98% relative). We showed that phoneme-level emotion models trained on "natural" emotions could provide impressive classification performance on dataset with acted affective content.

show abstract

“…During recognition the likelihood of generating the tested sequence of observations by a given HMM is calculated, and the emotion with the highest likelihood is selected. Monophone-based HMMs were also proposed for modeling frame-level acoustic features by Gajsek et al (2013), who achieved classification improvement both in emotion recognition and the alcohol detection, compared with the other state-of-the-art methods. Usually HMMs with Gaussian outputs are used.…”

Section: 4mentioning

confidence: 99%

Comparison of speaker dependent and speaker independent emotion recognition

Rybka

Janicki

2013

International Journal of Applied Mathematics and Computer Science

View full text Add to dashboard Cite

This paper describes a study of emotion recognition based on speech analysis. The introduction to the theory contains a review of emotion inventories used in various studies of emotion recognition as well as the speech corpora applied, methods of speech parametrization, and the most commonly employed classification algorithms. In the current study the EMO-DB speech corpus and three selected classifiers, the k-Nearest Neighbor (k-NN), the Artificial Neural Network (ANN) and Support Vector Machines (SVMs), were used in experiments. SVMs turned out to provide the best classification accuracy of 75.44% in the speaker dependent mode, that is, when speech samples from the same speaker were included in the training corpus. Various speaker dependent and speaker independent configurations were analyzed and compared. Emotion recognition in speaker dependent conditions usually yielded higher accuracy results than a similar but speaker independent configuration. The improvement was especially well observed if the base recognition ratio of a given speaker was low. Happiness and anger, as well as boredom and neutrality, proved to be the pairs of emotions most often confused.

show abstract

Speaker state recognition using an HMM-based feature extraction method

Cited by 25 publications

References 15 publications

Hidden MARKOV Model Approach Towards Emotion Detection From Speech Signal

Hidden MARKOV Model Approach Towards Emotion Detection From Speech Signal

Determining the Smallest Emotional Unit for Level of Arousal Classification

Comparison of speaker dependent and speaker independent emotion recognition

Contact Info

Product

Resources

About