.png){kind=icon}
返回
Wavelet Multiresolution Analysis Based Speech Emotion Recognition System Using 1D CNN LSTM Networks
.png){kind=icon}
OA
AI
DOI：10.1109/TASLP.2023.3277291.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Speech Emotion Recognition (SER) is the task of recognizing a speaker's emotional state from speech. SER plays a significant role in Human-Computer Interaction and psychological assessment. Several kinds of time-frequency representations like spectrograms, mel-frequency cepstrum coefficients (MFCCs), and mel-spectrograms are commonly used to develop an SER system. These representations use the Fast Fourier Transform (FFT) to convert the time domain signal to the frequency domain. However, the FFT has one fundamental limitation due to the uncertainty principle, which does not simultaneously allow a good resolution in both time and frequency domains. On the other hand, the multiresolution property of wavelets can provide a good localization in both time and frequency domains. Therefore, this article investigates the competency of the wavelet transforms for SER. We propose a Wavelet based Deep Emotion Recognition (WaDER) method using an autoencoder and 1D convolutional neural network (CNN) and long short-term memory (LSTM) networks. The autoencoder is used to perform the dimensionality reduction of the wavelet features then the latent space is used to classify the emotions using the 1D CNN-LSTM model. We conducted a Monte-Carlo K-fold validation using the Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) dataset. For speaker-dependent (SD) experiments, we achieved an unweighted accuracy (UA) of 81.45% and a weighted accuracy (WA) of 81.22%. The results of the experiments on the RAVDESS dataset show that the proposed method performs better than the state-of-the-art methods, which use other time-frequency representations.
Keyword:
Wavelet packets
Hidden Markov models
Convolutional neural networks
Continuous wavelet transforms
Time-frequency analysis
Feature extraction
Emotion recognition
Speech emotion recognition
wavelets
long short-term memory (LSTM)
convolutional neural networks (CNN)
autoencoders
dimensionality reduction
期刊
I
IF:
5.1
论文数:
2.6K
被引数:
1.1W
机构
引用论文
分享收藏A Historical Earthquake-Induced Landslide Damming Event at the Qiaojia Reach of the Jinsha River, SE Tibetan Plateau: Implication for the Seismic Hazard of the Xiaojiang Fault金沙江巧家段历史地震诱发滑坡堵江事件：对 Xiaojiang Fault 地震灾害的启示OAAI
OAAI分享收藏Long-term hydroclimatic variability over the semi-arid Ethiopian highlands in relation to ENSO and IOD teleconnection signals半干旱埃塞俄比亚高地的长期水气候变率及其与ENSO和IOD遥相关信号的关系OAAI
OAAI分享收藏Clustering-Based Speech Emotion Recognition by Incorporating Learned Features and Deep BiLSTM.png)
IEEE ACCESS
OAAI
IEEE ACCESSIF3.6
OAAI分享收藏
查看更多内容