Speech enhancement from fused features based on deep neural network and gated recurrent unit network

• Published in: EURASIP journal on advances in signal processing Vol. 2021; no. 1; pp. 1 - 19
• Main Authors: Wang, Youming, Han, Jiali, Zhang, Tianqi, Qing, Didi
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 24.10.2021; Springer; Springer Nature B.V; SpringerOpen
• Subjects: Algorithms; Analysis; Artificial neural networks; Computational linguistics; Context; Deep learning; Deep neural network; Engineering; Gated recurrent unit; Intelligibility; Language processing; Machine learning; Mapping; Natural language interfaces; Neural networks; Noise; Quantum Information Technology; Signal,Image and Speech Processing; Speech; Speech enhancement; Speech processing; Speech quality; Spintronics; Speech enhancement; Gated recurrent unit; Deep neural network; Speech quality
• ISSN: 1687-6180; 1687-6172; 1687-6180
• DOI: 10.1186/s13634-021-00813-8

Speech is easily interfered by external environment in reality, which results in the loss of important features. Deep learning has become a popular speech enhancement method because of its superior potential in solving nonlinear mapping problems for complex features. However, the deficiency of traditional deep learning methods is the weak learning capability of important information from previous time steps and long-term event dependencies between the time-series data. To overcome this problem, we propose a novel speech enhancement method based on the fused features of deep neural networks (DNNs) and gated recurrent unit (GRU). The proposed method uses GRU to reduce the number of parameters of DNNs and acquire the context information of the speech, which improves the enhanced speech quality and intelligibility. Firstly, DNN with multiple hidden layers is used to learn the mapping relationship between the logarithmic power spectrum (LPS) features of noisy speech and clean speech. Secondly, the LPS feature of the deep neural network is fused with the noisy speech as the input of GRU network to compensate the missing context information. Finally, GRU network is performed to learn the mapping relationship between LPS features and log power spectrum features of clean speech spectrum. The proposed model is experimentally compared with traditional speech enhancement models, including DNN, CNN, LSTM and GRU. Experimental results demonstrate that the PESQ, SSNR and STOI of the proposed algorithm are improved by 30.72%, 39.84% and 5.53%, respectively, compared with the noise signal under the condition of matched noise. Under the condition of unmatched noise, the PESQ and STOI of the algorithm are improved by 23.8% and 37.36%, respectively. The advantage of the proposed method is that it uses the key information of features to suppress noise in both matched and unmatched noise cases and the proposed method outperforms other common methods in speech enhancement.