Deep Learning based Emotion Recognition System Using Speech Features and Transcriptions

Published in

arXiv.org(2019)

External Links:

Generate Graph

TL;DR

The combined MFCC-Text Convolutional Neural Network model proved to be the most accurate in recognizing emotions in IEMOCAP data.

Abstract

This paper proposes a speech emotion recognition method based on speech features and speech transcriptions (text). Speech features such as Spectrogram and Mel-frequency Cepstral Coefficients (MFCC) help retain emotion-related low-level characteristics in speech whereas text helps capture semantic meaning, both of which help in different aspects of emotion detection. We experimented with several Deep Neural Network (DNN) architectures, which take in different combinations of speech features and text as inputs. The proposed network architectures achieve higher accuracies when compared to state-of-the-art methods on a benchmark dataset. The combined MFCC-Text Convolutional Neural Network (CNN) model proved to be the most accurate in recognizing emotions in IEMOCAP data.

Authors

Suraj Tripathi

1 papers

Abhay Kumar

1 papers

A. Ramesh

1 papers

References24 items

Multi-Modal Emotion recognition on IEMOCAP Dataset using Deep Learning

Efficient Emotion Recognition from Speech Using Deep Learning on Spectrograms

Tensor Fusion Network for Multimodal Sentiment Analysis

End-to-End Multimodal Emotion Recognition Using Deep Neural Networks

Emotion Recognition From Speech With Recurrent Neural Networks

Deep Learning based Emotion Recognition System Using Speech Features and Transcriptions

Published in

arXiv.org(2019)

External Links:

Generate Graph

TL;DR

The combined MFCC-Text Convolutional Neural Network model proved to be the most accurate in recognizing emotions in IEMOCAP data.

Abstract

Authors

Suraj Tripathi

1 papers

Abhay Kumar

1 papers

A. Ramesh

1 papers

References24 items

Multi-Modal Emotion recognition on IEMOCAP Dataset using Deep Learning

Efficient Emotion Recognition from Speech Using Deep Learning on Spectrograms

Tensor Fusion Network for Multimodal Sentiment Analysis

End-to-End Multimodal Emotion Recognition Using Deep Neural Networks

Emotion Recognition From Speech With Recurrent Neural Networks

Chirag Singh

1 papers

Promod Yenigalla

1 papers

Multimodal emotion recognition using deep learning architectures

Towards real-time Speech Emotion Recognition using deep neural networks

An experimental study of speech emotion recognition based on deep convolutional neural networks

Emotion recognition of affective speech based on multiple classifiers using acoustic-prosodic information and semantic labels (Extended abstract)

High-level feature representation using recurrent neural network for speech emotion recognition

Character-level Convolutional Networks for Text Classification

Speech emotion recognition with acoustic and lexical features

Speech emotion recognition using deep neural network and extreme learning machine

Convolutional Neural Networks for Sentence Classification

EEG-based emotion classification using deep belief networks

Deep learning for robust feature generation in audiovisual emotion recognition

Efficient Estimation of Word Representations in Vector Space

Deep Neural Networks for Acoustic Modeling in Speech Recognition

Emotion recognition in spontaneous speech using GMMs

Multi-level Speech Emotion Recognition Based on HMM and ANN

Fast and accurate sequential floating forward feature selection with the Bayes classifier applied to speech emotion recognition

IEMOCAP: interactive emotional dyadic motion capture database

GMM Supervector Based SVM with Spectral Features for Speech Emotion Recognition

Modeling the Temporal Evolution of Acoustic Parameters for Speech Emotion Recognition

Field of Study

Computer ScienceEngineeringMathematics

Journal Information

Name

ArXiv

Volume

abs/2005.00687

Venue Information

Name

arXiv.org

Type

URL

https://arxiv.org

Alternate Names

ArXiv