SPEECH DEREVERBERATION BASED ON INTEGRATED DEEP AND ENSEMBLE LEARNING ALGORITHM

Lee, Wei-Jen1; Wang, Syu-Siang1; Chen, Fei2; Lu, Xugang3; Chien, Shao-Yi4; Tsao, Yu1

10.1109/ICASSP.2018.8462662

2018

1520-6149

978-1-5386-4659-5

2018 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)

2018-April

5454-5458

15-20 April 2018

Calgary, AB, Canada

345 E 47TH ST, NEW YORK, NY 10017 USA

IEEE

Reverberation, which is generally caused by sound reflections from walls, ceilings, and floors, can result in severe performance degradation of acoustic applications. Due to a complicated combination of attenuation and time-delay effects, the reverberation property is difficult to characterize, and it remains a challenging task to effectively retrieve the anechoic speech signals from reverberation ones. In the present study, we proposed a novel integrated deep and ensemble learning algorithm (IDEA) for speech dereverberation. The IDEA consists of offline and online phases. In the offline phase, we train multiple dereverberation models, each aiming to precisely dereverb speech signals in a particular acoustic environment; then a unified fusion function is estimated that aims to integrate the information of multiple dereverberation models. In the online phase, an input utterance is first processed by each of the dereverberation models. The outputs of all models are integrated accordingly to generate the final anechoic signal. We evaluated the IDEA on designed acoustic environments, including both matched and mismatched conditions of the training and testing data. Experimental results confirm that the proposed IDEA outperforms single deep-neural-network-based dereverberation model with the same model architecture and training data.

Deep neural networks Speech dereverberation Ensemble learning Convolutional neural networks Deep denoising autoencoder

其他

英语

CPCI-S ; EI

Ministry of Science and Technology of Taiwan[MOST 107-2633-E-002-001]

Acoustics ; Engineering

Acoustics ; Engineering, Electrical & Electronic

WOS:000446384605125

20184005907190

Web of Science

1.Acad Sinica, Res Ctr Informat Technol Innovat, Taipei, Taiwan
2.Southern Univ Sci & Technol, Dept Elect & Elect Engn, Shenzhen, Peoples R China
3.Natl Inst Informat & Commun Technol, Tokyo, Japan
4.Natl Taiwan Univ, Dept Elect Engn, Taipei, Taiwan

Lee, Wei-Jen,Wang, Syu-Siang,Chen, Fei,et al. SPEECH DEREVERBERATION BASED ON INTEGRATED DEEP AND ENSEMBLE LEARNING ALGORITHM[C]. 345 E 47TH ST, NEW YORK, NY 10017 USA:IEEE,2018:5454-5458.