Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Research › peer-review
CNN based Emotion Detection in Cross Linguistic Children speech. / Lyakso, Elena ; Frolova, Olga ; Ruban, Nersissona; Mekala A., Mary; Noel, Joseph Raj, Alex ; Matveev, Anton ; Matveev, Yuri.
Proceedings of the 24 th International Congress of Acoustics: A15 SPEECH. Vol. A15 Gyeongju, 2022. p. 8-14 ABS-0435.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Research › peer-review
}
TY - GEN
T1 - CNN based Emotion Detection in Cross Linguistic Children speech
AU - Lyakso, Elena
AU - Frolova, Olga
AU - Ruban, Nersissona
AU - Mekala A., Mary
AU - Noel, Joseph Raj, Alex
AU - Matveev, Anton
AU - Matveev, Yuri
PY - 2022
Y1 - 2022
N2 - The goal of this cross-linguistic study is to analyze the speech of children in the age group of 8-12 years andautomatic classification of children's emotions from various speech data in Russian and Tamil languages. Astandardized approach for the collection of speech (spontaneous and acting speech - emotional words,phrases, and meaningless texts) was used. Two emotional child speech corpora include the emotional speechof 95 Russian children and 40 Indian children were created. Annotation of the emotional speech was carriedout in four categories “joy - neutral - sadness – anger” by two speech specialists of the same nationality of thechild. The set of 2505 labeled audio files of Russian children and 418 labeled audio files of Tamil childrenwere used. The SVM classifier shows slightly better results in Russian language, and the MLP classifier inTamil. Inter-Cultural approach (mixed dataset of Russian and Indian speech) revealed that the accuracy ofrecognition of all emotions remains higher. In Cross-Cultural approach, on samples of Tamil speech, angerstate is recognized better, and in samples of Russian speech – sad state. Using 5 Layered CNN Model, theaccuracy was revealed for Russian speech is higher than for Indian children.
AB - The goal of this cross-linguistic study is to analyze the speech of children in the age group of 8-12 years andautomatic classification of children's emotions from various speech data in Russian and Tamil languages. Astandardized approach for the collection of speech (spontaneous and acting speech - emotional words,phrases, and meaningless texts) was used. Two emotional child speech corpora include the emotional speechof 95 Russian children and 40 Indian children were created. Annotation of the emotional speech was carriedout in four categories “joy - neutral - sadness – anger” by two speech specialists of the same nationality of thechild. The set of 2505 labeled audio files of Russian children and 418 labeled audio files of Tamil childrenwere used. The SVM classifier shows slightly better results in Russian language, and the MLP classifier inTamil. Inter-Cultural approach (mixed dataset of Russian and Indian speech) revealed that the accuracy ofrecognition of all emotions remains higher. In Cross-Cultural approach, on samples of Tamil speech, angerstate is recognized better, and in samples of Russian speech – sad state. Using 5 Layered CNN Model, theaccuracy was revealed for Russian speech is higher than for Indian children.
KW - child speech
KW - Emotion Detection
KW - Cross-Linguistic Study
UR - https://www.ica2022korea.org/data/Proceedings_A15.pdf
M3 - Conference contribution
VL - A15
SP - 8
EP - 14
BT - Proceedings of the 24 th International Congress of Acoustics
CY - Gyeongju
Y2 - 24 October 2022 through 28 November 2022
ER -
ID: 100861799