Speech Emotion Recognition using Enriched Spectrogram and Deep Convolutional Neural Network Transfer Learning
Publish Year: 1401
نوع سند: مقاله ژورنالی
زبان: English
View: 242
This Paper With 10 Page And PDF Format Ready To Download
- Certificate
- من نویسنده این مقاله هستم
این Paper در بخشهای موضوعی زیر دسته بندی شده است:
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
JR_JADM-10-4_008
تاریخ نمایه سازی: 28 آذر 1401
Abstract:
Speech emotion recognition (SER) is a challenging field of research that has attracted attention during the last two decades. Feature extraction has been reported as the most challenging issue in SER systems. Deep neural networks could partially solve this problem in some other applications. In order to address this problem, we proposed a novel enriched spectrogram calculated based on the fusion of wide-band and narrow-band spectrograms. The proposed spectrogram benefited from both high temporal and spectral resolution. Then we applied the resultant spectrogram images to the pre-trained deep convolutional neural network, ResNet۱۵۲. Instead of the last layer of ResNet۱۵۲, we added five additional layers to adopt the model to the present task. All the experiments performed on the popular EmoDB dataset are based on leaving one speaker out of a technique that guarantees the speaker's independency from the model. The model gains an accuracy rate of ۸۸.۹۷% which shows the efficiency of the proposed approach in contrast to other state-of-the-art methods.
Keywords:
Wideband and narrowband spectrogram , ResNet۱۵۲ , DCNN , Transfer learning , Speech emotion recognition
Authors
B. Z. Mansouri
Electrical and Computer Engineering Department, Ferdows branch, Islamic Azad University, Ferdows, Iran.
H.R. Ghaffary
Electrical and Computer Engineering Department, Ferdows branch, Islamic Azad University, Ferdows, Iran.
A. Harimi
Electrical and Computer Engineering Department, Ferdows branch, Islamic Azad University, Ferdows, Iran.
مراجع و منابع این Paper:
لیست زیر مراجع و منابع استفاده شده در این Paper را نمایش می دهد. این مراجع به صورت کاملا ماشینی و بر اساس هوش مصنوعی استخراج شده اند و لذا ممکن است دارای اشکالاتی باشند که به مرور زمان دقت استخراج این محتوا افزایش می یابد. مراجعی که مقالات مربوط به آنها در سیویلیکا نمایه شده و پیدا شده اند، به خود Paper لینک شده اند :