Progress in speech emotion recognition

Zhang, Xueying; Sun, Ying; Duan, Shufei

doi:10.1109/tencon.2015.7372840

Cited by 17 publications

(10 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• Time-Based Features They are zero-crossing rate (ZCR) [8] and amplitudebased features, such as amplitude descriptor, log attack time, attach, delay, sustain, release envelop, short-time energy (STE) [9], shimmer [10], rhythm-based features [8], [11], volume, and temporal centroid [3].…”

Section: Acoustic Features In Ser Literaturementioning

confidence: 99%

“…Spectral features that well simulate the vocal tract [7], [15]. The quality of the voice depends on the physical structure of the vocal tract and the glottal waveforms [8]. It includes HNR, jitter, shimmer, normalized amplitude quotient, quasi open quotient, parabolic spectral parameter, and maxima dispersion quotient [7], [15].…”

Section: Acoustic Features In Ser Literaturementioning

confidence: 99%

See 1 more Smart Citation

A Speech Emotion Recognition Model Based on Multi-Level Local Binary and Local Ternary Patterns

Sönmez

Varol

2020

IEEE Access

View full text Add to dashboard Cite

Interpreting a speech signal is quite challenging because it consists of different frequencies and features that vary according to emotions. Although different algorithms are being developed in the speech emotion recognition (SER) domain, the success rates vary according to the spoken languages, emotions, and databases. In this study, a new lightweight effective SER method has been developed that has low computational complexity. This method, called 1BTPDN, is applied on RAVDESS, EMO-DB, SAVEE, and EMOVO databases. First, low-pass filter coefficients are obtained by applying a one-dimensional discrete wavelet transform on the raw audio data. The features are extracted by applying textural analysis methods, a one-dimensional local binary pattern, and a one-dimensional local ternary pattern to each filter. Using neighborhood component analysis, the most dominant 1024 features are selected from 7680 features while the other features are discarded. These 1024 features are selected as the input of the classifier which is a thirddegree polynomial kernel-based support vector machine. The success rates of the 1BTPDN reached 95.16%, 89.16%, 76.67%, and 74.31% in the RAVDESS, EMO-DB, SAVEE, and EMOVO databases, respectively. The recognition rates are higher compared to many textural, acoustic, and deep learning state-of-the-art SER methods. INDEX TERMS Discrete wavelet transform, local binary pattern, local ternary pattern, neighborhood component analysis, speech emotion recognition.

show abstract

Section: Acoustic Features In Ser Literaturementioning

confidence: 99%

Section: Acoustic Features In Ser Literaturementioning

confidence: 99%

A Speech Emotion Recognition Model Based on Multi-Level Local Binary and Local Ternary Patterns

Sönmez

Varol

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Most of the modern investigations report speech emotion classification rates of 70-90% [4,45,54]. Particular results depend on the analyzed language, the number of emotions, the speaker mode, and other important factors.…”

Section: Introductionmentioning

confidence: 99%

Speech emotion classification using fractal dimension-based features

Tamulevičius

Karbauskaitė

Dzemyda

2019

NAMC

View full text Add to dashboard Cite

During the last 10–20 years, a great deal of new ideas have been proposed to improve the accuracy of speech emotion recognition: e.g., effective feature sets, complex classification schemes, and multi-modal data acquisition. Nevertheless, speech emotion recognition is still the task in limited success. Considering the nonlinear and fluctuating nature of the emotional speech, in this paper, we present fractal dimension-based features for speech emotion classification. We employed Katz, Castiglioni, Higuchi, and Hurst exponent-based features and their statistical functionals to establish the 224-dimensional full feature set. The dimension was downsized by applying the Sequential Forward Selection technique. The results of experimental study show a clear superiority of fractal dimension-based feature sets against the acoustic ones. The average accuracy of 96.5% was obtained using the reduced feature sets. The feature selection enabled us to obtain the 4-dimensional and 8-dimensional sets for Lithuanian and German emotions, respectively.

show abstract

“…Speaker emotion recognition is based on the emotional feature analysis. The most important speech emotion features include speech quality feature, spectral feature and prosodic feature [1,2]. We cannot achieve good results using only one of these types of features.…”

Section: Introductionmentioning

confidence: 99%

“…emotion feature number). Then each utterance feature ijn P ( n is the emotion sample index) is normalized Calculate the feature dispersion for certain emotion1 The feature contribution and the Euclidean distance are summed ( ) each feature is taken as the sum weight and it is simple enough for KNN classification. The inner relation between features is modeled and the performance of emotion classification is improved.…”

mentioning

confidence: 99%

Speech Emotion Recognition Based on Fuzzy K-NN Algorithm with Fractionally Spaced Blind Equalization

Yuan¹,

Deng²,

Shi³

2016

Proceedings of the 2016 2nd Workshop on Advanced Research and Technology in Industry Applications

View full text Add to dashboard Cite

Due to the noise and multi-channel room acoustic environment, practical speech emotion recognition remains an unsolved challenge. In this paper, we first study the fractional spaced blind equalizer for speech preprocessing. The noise interference is effectively removed and more detailed emotional features are reserved. Second, the fuzzy k nearest neighbor algorithm is used to classify speech. Finally, the proposed algorithm is compared with traditional speech emotion recognition algorithms. Experimental results show that the fractionally spaced equalization is effective for practical speech emotion recognition.

show abstract

Progress in speech emotion recognition

Cited by 17 publications

References 41 publications

A Speech Emotion Recognition Model Based on Multi-Level Local Binary and Local Ternary Patterns

A Speech Emotion Recognition Model Based on Multi-Level Local Binary and Local Ternary Patterns

Speech emotion classification using fractal dimension-based features

Speech Emotion Recognition Based on Fuzzy K-NN Algorithm with Fractionally Spaced Blind Equalization

Contact Info

Product

Resources

About