Emotion modeling from speech signal based on wavelet packet transform

Degaonkar, Varsha; Apte, Shaila Dinkar

doi:10.1007/s10772-012-9142-8

Cited by 16 publications

(9 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3. At the beginning the speech signals of all speakers were normalized in reference to the square root of energy, according to the formula (9) Then, for each normalized recording, DFWT were computed, as it was described in Section II. …”

Section: Methodsmentioning

confidence: 99%

“…In another approach [8], experiments on applicability of DWT, Wavelet Packets Transform (WPT) and Perceptual Wavelet Packet Transform (PWPT) are performed, where the best results were achieved for the DWT method. Emotion modelling and emotion conversion using transition in subbands of WPT is performed in [9]. The fusion of Fourier and DWT has been investigated in the realm of emotion detection and some results are presented in this paper.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Combination of Fourier and wavelet transformations for detection of speech emotions

Ziółko

Jaciow

Igras

2014

2014 7th International Conference on Human System Interactions (HSI)

View full text Add to dashboard Cite

The paper presents an approach to automatic recognition of emotions in speech signals. The applied method bases on the composition of two discrete frequency transformations. The wavelet transform was calculated first and next the Fourier transform was applied. The Fourier-wavelet transform representation is used to find the differences between emotions in speech signals. A set of approximately 30 seconds long speech signals was used to verify the efficiency of presented methods. It gives the possibility of analyzing the performance of speech emotion recognition in the Fourier-wavelet domain.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Combination of Fourier and wavelet transformations for detection of speech emotions

Ziółko

Jaciow

Igras

2014

2014 7th International Conference on Human System Interactions (HSI)

View full text Add to dashboard Cite

show abstract

“…The final design of the model fit to calculate the mean pitch model for Marathi emotional speech calculated as shown in equation ( 7) as below, Mean pitch = Emotion + Gender + (1/speaker) (7) As in equation ( 7), emotion and gender factors are fixed effect variables, and the speaker is a random effect variable for calculating the mean pitch model for Marathi's emotional speech. Table VI shows the impacts of fixed effect variables such as emotions (angry, happiness, fear, and neutral) and gender (male and female) on computing mean pitch values.…”

Section: ) Modeling Mean Pitchmentioning

confidence: 99%

“…Acoustically, the boundaries showed preboundary lengthening and pitch contour slope on the final syllable, and the prominence correlated with maximum F0 and maximum intensity and lesser duration. The authors analyzed MFCC features and energy ratios in [7] to investigate the Marathi emotion recognition for anger and happiness. The authors observed the anger emotion recognition rate higher than happiness and neutral emotions.…”

Section: Introductionmentioning

confidence: 99%

Linear Mixed Effect Modelling for Analyzing Prosodic Parameters for Marathi Language Emotions

Harhare¹,

Shah²

2021

IJACSA

View full text Add to dashboard Cite

Along with linguistic messages, prosody is an essential paralinguistic component of emotional speech. Prosodic parameters such as intensity, fundamental frequency (F0), and duration were studied worldwide to understand the relationship between emotions and corresponding prosody features for various languages. For evaluating prosodic aspects of emotional Marathi speech, the Marathi language has received less attention. This study aims to see how different emotions affect suprasegmental properties such as pitch, duration, and intensity in Marathi's emotional speech. This study investigates the changes in prosodic features based on emotions, gender, speakers, utterances, and other aspects using a database with 440 utterances in happiness, fear, anger, and neutral emotions recorded by eleven Marathi professional artists in a recording studio. The acoustic analysis of the prosodic features was employed using PRAAT, a speech analysis framework. A statistical study using a two-way Analysis of Variance (two-way ANOVA) explores emotion, gender, and their interaction for mean pitch, mean intensity, and sentence utterance time. In addition, three distinct linear mixed-effect models (LMM), one for each prosody characteristic designed comprising emotion and gender factors as fixed effect variables, whereas speakers and sentences as random effect variables. The relevance of the fixed effect and random effect on each prosodic variable was verified using likelihood ratio tests that assess the goodness of fit. Based on Marathi's emotional speech, the R programming language examined linear mixed modeling for mean pitch, mean intensity, and sentence duration.

show abstract

“…for emotion recognition in human interactions. There are many methods for SER, those are Neural Network [7], wavelet Packet Transform [8], Support Vector Machine [9], etc. The existing SER technique have several issues such as a number of unknown features increase the computational complexity [10], noisy data, language-dependent [11], presence of irrelevant data, curse of dimensionality [12].…”

Section: Introductionmentioning

confidence: 99%

Hybrid Deep Network Scheme for Emotion Recognition in Speech

Angadi¹,

Reddy²

2019

IJIES

View full text Add to dashboard Cite

Speech Emotion Recognition (SER) is an active research area with wide range of applications like medical, entertainment, monitoring in the field of Human Computer Interface. Generally, speech signals include high dimensionality of the feature that degrades the performance of SER. This research paper concentrated on SER using hybrid network that is composed of Convolutional Neural Network and Bidirectional Long Short Term Memory Networks (CNN-BLSTM). The major objective of this study is to recognize more relevant temporal features rather than traditional feature learning. The proposed CNN-BLSTM techniques increases the precision, recall and accuracy of emotion recognition in speech signal significantly. IEMOCAP dataset has been used in experimental analysis of the proposed approach to classify the different emotions of human; those are happy, angry, sad, and neutral. The performance of the CNN-BLSTM method has been measured using parameters like precision, recall and accuracy. Compare to the existing Support Vector Machine (SVM), the proposed CNN-BLSTM achieved approximately 9.83% of true positives proportion enhancement in SER.

show abstract

Emotion modeling from speech signal based on wavelet packet transform

Cited by 16 publications

References 1 publication

Combination of Fourier and wavelet transformations for detection of speech emotions

Combination of Fourier and wavelet transformations for detection of speech emotions

Linear Mixed Effect Modelling for Analyzing Prosodic Parameters for Marathi Language Emotions

Hybrid Deep Network Scheme for Emotion Recognition in Speech

Contact Info

Product

Resources

About