IEEE

Accuracy Enhancement Method for Speech Emotion Recognition From Spectrogram Using Temporal Frequency Correlation and Positional Information Learning Through Knowledge Transfer

Abstract: In this paper, we propose a method to improve the accuracy of speech emotion recognition (SER) by using vision transformer (ViT) to attend to the correlation of frequency (y-axis) with time ...
IEEE

RecSpecUNet: A Recycled and Selective Architecture for Efficient Spectrogram Segmentation

Abstract: This paper presents RecSpecUNet, a deep learning model based on UNet, designed for automatic detection of Fifth Generation (5G), Long-Term Evolution (LTE), and radar signals in ...