Extraction of Novel Features Based on Histograms of MFCCs Used in Emotion Classification from Generated Original Speech Dataset
This paper introduces two significant contributions: one is a new feature based on histograms of MFCC (Mel-Frequency Cepstral Coefficients) extracted from the audio files that can be used in emotion classification from speech signals, and the other – our new multi-lingual and multi-personal speech d...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Kaunas University of Technology
2020-02-01
|
| Series: | Elektronika ir Elektrotechnika |
| Subjects: | |
| Online Access: | http://eejournal.ktu.lt/index.php/elt/article/view/25309 |
| _version_ | 1818340479596494848 |
|---|---|
| author | Muhammet Pakyurek Mahir Atmis Selman Kulac Umut Uludag |
| author_facet | Muhammet Pakyurek Mahir Atmis Selman Kulac Umut Uludag |
| author_sort | Muhammet Pakyurek |
| collection | DOAJ |
| description | This paper introduces two significant contributions: one is a new feature based on histograms of MFCC (Mel-Frequency Cepstral Coefficients) extracted from the audio files that can be used in emotion classification from speech signals, and the other – our new multi-lingual and multi-personal speech database, which has three emotions. In this study, Berlin Database (BD) (in German) and our custom PAU database (in English) created from YouTube videos and popular TV shows are employed to train and evaluate the test results. Experimental results show that our proposed features lead to better classification of results than the current state-of-the-art approaches with Support Vector Machine (SVM) from the literature. Thanks to our novel feature, this study can outperform a number of MFCC features and SVM classifier based studies, including recent researches. Due to the lack of our novel feature based approaches, one of the most common MFCC and SVM framework is implemented and one of the most common database Berlin DB is used to compare our novel approach with these kind of approaches. |
| first_indexed | 2024-12-13T15:43:33Z |
| format | Article |
| id | doaj.art-fa5966e2c78141c3a628061378ac035d |
| institution | Directory Open Access Journal |
| issn | 1392-1215 2029-5731 |
| language | English |
| last_indexed | 2024-12-13T15:43:33Z |
| publishDate | 2020-02-01 |
| publisher | Kaunas University of Technology |
| record_format | Article |
| series | Elektronika ir Elektrotechnika |
| spelling | doaj.art-fa5966e2c78141c3a628061378ac035d2022-12-21T23:39:46ZengKaunas University of TechnologyElektronika ir Elektrotechnika1392-12152029-57312020-02-01261465110.5755/j01.eie.26.1.2530925309Extraction of Novel Features Based on Histograms of MFCCs Used in Emotion Classification from Generated Original Speech DatasetMuhammet PakyurekMahir AtmisSelman KulacUmut UludagThis paper introduces two significant contributions: one is a new feature based on histograms of MFCC (Mel-Frequency Cepstral Coefficients) extracted from the audio files that can be used in emotion classification from speech signals, and the other – our new multi-lingual and multi-personal speech database, which has three emotions. In this study, Berlin Database (BD) (in German) and our custom PAU database (in English) created from YouTube videos and popular TV shows are employed to train and evaluate the test results. Experimental results show that our proposed features lead to better classification of results than the current state-of-the-art approaches with Support Vector Machine (SVM) from the literature. Thanks to our novel feature, this study can outperform a number of MFCC features and SVM classifier based studies, including recent researches. Due to the lack of our novel feature based approaches, one of the most common MFCC and SVM framework is implemented and one of the most common database Berlin DB is used to compare our novel approach with these kind of approaches.http://eejournal.ktu.lt/index.php/elt/article/view/25309emotion classificationmfccsvmspeech signal |
| spellingShingle | Muhammet Pakyurek Mahir Atmis Selman Kulac Umut Uludag Extraction of Novel Features Based on Histograms of MFCCs Used in Emotion Classification from Generated Original Speech Dataset Elektronika ir Elektrotechnika emotion classification mfcc svm speech signal |
| title | Extraction of Novel Features Based on Histograms of MFCCs Used in Emotion Classification from Generated Original Speech Dataset |
| title_full | Extraction of Novel Features Based on Histograms of MFCCs Used in Emotion Classification from Generated Original Speech Dataset |
| title_fullStr | Extraction of Novel Features Based on Histograms of MFCCs Used in Emotion Classification from Generated Original Speech Dataset |
| title_full_unstemmed | Extraction of Novel Features Based on Histograms of MFCCs Used in Emotion Classification from Generated Original Speech Dataset |
| title_short | Extraction of Novel Features Based on Histograms of MFCCs Used in Emotion Classification from Generated Original Speech Dataset |
| title_sort | extraction of novel features based on histograms of mfccs used in emotion classification from generated original speech dataset |
| topic | emotion classification mfcc svm speech signal |
| url | http://eejournal.ktu.lt/index.php/elt/article/view/25309 |
| work_keys_str_mv | AT muhammetpakyurek extractionofnovelfeaturesbasedonhistogramsofmfccsusedinemotionclassificationfromgeneratedoriginalspeechdataset AT mahiratmis extractionofnovelfeaturesbasedonhistogramsofmfccsusedinemotionclassificationfromgeneratedoriginalspeechdataset AT selmankulac extractionofnovelfeaturesbasedonhistogramsofmfccsusedinemotionclassificationfromgeneratedoriginalspeechdataset AT umutuludag extractionofnovelfeaturesbasedonhistogramsofmfccsusedinemotionclassificationfromgeneratedoriginalspeechdataset |