October 2024
What is this page?
This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.
Publications (23)
October 2024
·
3 Reads
August 2024
·
7 Reads
November 2022
·
3 Reads
·
1 Citation
November 2022
·
1 Read
·
1 Citation
November 2021
·
20 Reads
·
4 Citations
July 2019
·
20 Reads
·
1 Citation
July 2019
·
19 Reads
·
3 Citations
July 2017
·
117 Reads
·
1 Citation
June 2017
·
32 Reads
·
1 Citation
Citations (12)
... For instance, for the EmoDB corpus, results with deep learning based classifiers range from 71% [17] to 95.89% [18]. Since we focus our efforts on the Romanian language, the efforts of Feraru and Zbancionc [10,19] are worth mentioning. Using a deep learning approach and the Emo-IIT corpus, they obtained approximately 85% accuracy, computed using 10-fold crossvalidation. ...
- Citing Conference Paper
November 2021
... Our Bengali dataset consists of around 10,500 Bengali voice clips to test our system. First, our dataset is designed in text format encompassing all required states of emotion [21]. Then, recorded most of the voice clips on our own. ...
- Citing Conference Paper
July 2019
... Physical states include various nonverbal cues, like utterance rate [7,8], nodding [9][10][11], and smiling [12,13], which provides insights into an individual's behavioral expressions. Cognitive states, on the other hand, comprise complex mental processes, such as engagement [14][15][16], boredom [17][18][19], and self-confidence [20][21][22], which are essential for understanding an individual's mental and emotional states. ...
- Citing Conference Paper
July 2017
... Consequently, feature sets or models that perform well on one dataset may translate poorly to others, particularly when there are disparities in data generation methodologies 14 . As a result, research involving crosscorpus evaluations of depression frequently produces performance levels that approach random chance when applied to mismatched datasets [15][16][17] . Unfortunately, this critical issue has not been thoroughly investigated in existing deep-learning models focused on speech-based depression recognition, highlighting a significant gap that requires further exploration. ...
- Citing Conference Paper
September 2015
... (1) In the study regarding the automatic recognition of anxiety emotional state using the EMO-DB dataset by using KNN [23], the proposed method achieved an average accuracy of 90.24%. Te accuracy for identifying anxiety/fear emotion specifcally was 89%. ...
- Citing Conference Paper
November 2015
... LDCs and kNN classifiers have been used since the very first studies and turned out to be quite successful for both acted and spontaneous emotional speech [19]. Zbancioc et al. used a weighted kNN classifier for the classification task of four emotions (anger, sadness, joy and neutral) contained in the SROL emotion corpus utilized in their research [36]. ...
- Citing Article
- Full-text available
August 2012
Advances in Electrical and Computer Engineering
... Tobin et al. [2] introduced the eigenvector weighting factor into the membership function, obtained an improved method for calculating semiconductor membership degree and fuzzy classification, in addition, applied it to the field of industrial automation. Zbancioc [3] and Choksi [4] et al. have applied fuzzy KNN to image recognition, and have made some explorations. Josien and Liao [5] combining fuzzy C-means and fuzzy KNN for GT part family and machine cell formation. ...
- Citing Conference Paper
November 2012
... The results showed that about 93.8% of syllables are detected with less than 20% of average syllable duration error. A predictive neural network for syllable nucleus prediction of Romanian language based on the extraction of prosodic parameters such as LPCC "Linear Prediction Cpestral Coefficients" and MFCC from the vowel areas was presented [15]. They used the SROL "Sounds of the Romanian Language" database [16] with a minimum resulting prediction error of 3%. ...
- Citing Conference Paper
July 2013
... Comparative experiments showed that this optimization method improved the accuracy by 4%. In [114], the authors measured the emotion recognition accuracy when LPC coefficients were introduced in the feature vectors. Using only the LOC coefficients, the model achieved 78% accuracy on the SROL dataset. ...
- Citing Conference Paper
November 2013
... NOTICES 1. A partial version [6] of this paper was presented in the ICVL 2009 conference and received the INTEL Special Award for Education (2009). 2. The authors contributions: the gnathophonic and gnathosonic research was been performed by the second author who also wrote the corresponding section of the paper (Sections 2, 4, 5, 6, and 8, and contributed to writing the other sections); the first author helped with further recordings and with their inclusion on the web page. ...
- Citing Article