Learning utterance-level representations for speech emotion and age/gender recognition using deep neural networks

Zhong-qiu Wang, Ivan Tashev. Learning utterance-level representations for speech emotion and age/gender recognition using deep neural networks. In 2017 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2017, New Orleans, LA, USA, March 5-9, 2017. pages 5150-5154, IEEE, 2017. [doi]

Authors

Zhong-qiu Wang

This author has not been identified. Look up 'Zhong-qiu Wang' in Google

Ivan Tashev

This author has not been identified. Look up 'Ivan Tashev' in Google