Speech Emotion Classification with Data Augmentation using Conditional Generative Adversial Network
Yikai Yang, Son Tran, Manisha Pratihast, Craig Jin
Sound Examples (Our Methods&Recorded&CWaveGAN)
Artificial 'Agree' Recorded 'Agree'. CWaveGAN 'Agree'
Artificial 'Bored' Recorded 'Bored' CWaveGAN 'Bored'
Artificial 'Confused' Recorded 'Confused'. CWaveGAN 'Confused'
Artificial 'Fear' Recorded 'Fear'. CWaveGAN 'Fear''
Artificial 'Happy' Recorded 'Happy' CWaveGAN 'Happy'
Artificial 'Neutural' Recorded 'Neutural' CWaveGAN 'Neutural''
Artificial 'Sad' Recorded 'Sad' CWaveGAN 'Sad'
We also did test on Ravdess (public dataset with long sentences), the demos will come together with our future journal paper