Zhen Huang

TL;DR: In this paper, a multi-objective framework is proposed to learn both secondary targets not directly related to the intended task of speech enhancement, and the primary target of the clean log-power spectra (LPS) features to be used directly for constructing the enhanced speech signals.

TL;DR: A series of experiments show that joint LPS and MFCC learning improves the SE performance, and IBM-based post-processing further enhances listening quality of the reconstructed speech.

TL;DR: The proposed MTL adaptation framework improves the learning ability of the original DNN structure, then enlarge the coverage of the acoustic space to deal with the unseen senone problem, and thus enhance the discrimination power of the adapted DNN models.

TL;DR: This paper proposes a novel method to detect and locate power outages based on the information collected from social media using Twitter as a real-time social sensor with a supervised topic model with a heterogeneous information network.

TL;DR: An integrated end-to-end automatic speech recognition (ASR) paradigm by joint learning of the front-end speech signal processing and back-end acoustic modeling is proposed, leading to a unified deep neural network (DNN) framework for distant speech processing that can achieve both high-quality enhanced speech and high-accuracy ASR simultaneously.