Sound based human emotion recognition using MFCC &amp; multiple SVM

Sonawane, Anagha; Inamdar, Madhura; Bhangale, Kishor

doi:10.1109/icomicon.2017.8279046

Cited by 34 publications

(13 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In study [9], the authors have leveraged MFCC for extraction of features and multiple Support Vector Machine (SVM) as a number of classifiers. Their extensive experiments are based on happiness, anger, sadness, disgust, surprise, and neutral emotion sound database.…”

Section: Extraction Of Novel Features Based On Histograms Of Mfccs Usmentioning

confidence: 99%

Extraction of Novel Features Based on Histograms of MFCCs Used in Emotion Classification from Generated Original Speech Dataset

Pakyurek¹,

Atmis²,

Kulac³

et al. 2020

ELEKTRON ELEKTROTECH

View full text Add to dashboard Cite

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.

show abstract

Section: Extraction Of Novel Features Based On Histograms Of Mfccs Usmentioning

confidence: 99%

Extraction of Novel Features Based on Histograms of MFCCs Used in Emotion Classification from Generated Original Speech Dataset

Pakyurek¹,

Atmis²,

Kulac³

et al. 2020

ELEKTRON ELEKTROTECH

View full text Add to dashboard Cite

show abstract

“…The speech emotion recognition system has used various processes and some papers have been introduced here. Anagha Sonawane et al [2] have shown a method for emotion recognition which consist of MFCC based speech signal features extraction. After this, features are classified according to labels using Multiple Support Vector Machine (SVM).…”

Section: Related Workmentioning

confidence: 99%

“…We tend to are completely focused on speech signal, during this paper varied reasonably options that carries data regarding message, speaker, language and emotion. The approach is to calculate numerous options that carries a lot of data and so mix these features to search out a strong recognition rate [2]. There are many applications in speech emotion recognition its helpful in healthcare center for detective work frustration.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Speech based Emotion Recognition using various Features and SVM Classifier

Jagtap¹

2019

IJRASET

View full text Add to dashboard Cite

In this paper methodology for human emotion recognizes by extracting the speech signal. This speaker-based emotion recognition system recognizes the four emotions namely happiness, sadness, fear and angry. Basically, aim of this system to recognize the emotions and estimate the various features namely formant frequency, energy, pitch and MFCC from speech signal. accuracy of emotion detection system using speech signal depends on types of feature used to extract unique characteristics in case of individual emotion recognition. emotions are showing effective changes in specific features and hence SVM (Support Vector Machine) classifier can show better performance. Individual emotion classification shows better performance in our work but in case of average it is considerably less i.e. 68%.

show abstract

“…[1]The approach is to calculate numerous options that carry a lot of data and identify the features to search out a strong recognition rate. [2] Speech based analysis to recognize is the best way to understand the actual intensions of the speaker. This programme developed is helpful to identify the various errors that an individual adapts due to the influence of regional language over command of the English language.…”

Section: Introductionmentioning

confidence: 99%

Speech Analysis based Pronunciation Correction Training for School Student

Tol¹

2019

IJRASET

View full text Add to dashboard Cite

Pronunciation accuracy enhancement is the need of education for school students during career development. The correct speech with exact pronunciation is what one can learn by understanding the phonetics definitions available in various dictionaries and study material of the English language. There is a need for a system for teaching the pronunciation, and that varies when run by humans based on individual teacher's performance. Also, due to variations in language, as practically every 25 kilometers of the span is responsible for variation in language, a computerized system for pronunciation learning can provide a solution to the problem. In this paper, we have provided speech analysis based tool for pronunciation accuracy improvement system which makes use of Google speech API for text to speech conversion. The speech of the user is recorded using a microphone. The recorded speech is analyzed based on MFCC features and compared with that of the speech obtained from google speech API. The improvement for correct pronunciation is seen amongst students when they are provided the tool for experimentation as comparison is obtained in terms of percentile score along with speech recognition and then converting it into the text from improper to proper speech scenarios. The trials are performed on more than 200 school students which shows satisfactory results.

show abstract

Sound based human emotion recognition using MFCC & multiple SVM

Cited by 34 publications

References 6 publications

Extraction of Novel Features Based on Histograms of MFCCs Used in Emotion Classification from Generated Original Speech Dataset

Extraction of Novel Features Based on Histograms of MFCCs Used in Emotion Classification from Generated Original Speech Dataset

Speech based Emotion Recognition using various Features and SVM Classifier

Speech Analysis based Pronunciation Correction Training for School Student

Contact Info

Product

Resources

About