Yue Xie

Dr. Yue Xie


Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 March 2024 in IEEE/ACM Transactions on Audio, Speech, and Language Processing
Verified
Journal Article
Closed access
RLRuiyu Liang
Ruiyu Liang
YXYue Xie
Yue Xie
JCJiaming Cheng
Jiaming Cheng
CPCong Pang
Cong Pang
BSBjörn Schuller
Björn Schuller
The speech quality delivered by hearing aids plays a crucial role in determining the acceptance and satisfaction of users. Compared with invasive speech quality evaluation methods that require pure signals as a reference, this paper proposes a non-invasive speech quality evaluation algorithm for hearing aids with multi-head self-attention and audiogram-based features. Initially, the audiogram of hearing-impaired individuals is extended along the frequency axis, enabling the speech quality evaluation model to learn the gain requirements specific to frequency bands for hearing-impaired individuals. Subsequently, the spectrogram is extracted from the speech signals to be evaluated. These features are combined with the transformed audiogram to create input features. To extract deep frame-level feature, a network employing multiple two-dimensional convolutional modules is utilized. Then, the temporal features are modeled using bidirectional long short-term memory networks (BiLSTM), while a multi-head self-attention mechanism is employed to integrate contextual information. This mechanism enables the model to focus on key frame information. Experimental results demonstrate that, compared to currently available advanced algorithms, the proposed network exhibits a higher correlation with the Hearing Aid Speech Quality Index (HASQI) and demonstrates robustness under various noise conditions.

1 May 2023 in IEICE Transactions on Information and Systems
Verified
Journal Article
Open access
PDF file
YXYue Xie
Yue Xie
RLRuiyu Liang
Ruiyu Liang
ZLZhenlin Liang
Zhenlin Liang
XZXiaoyan Zhao
Xiaoyan Zhao
WZWenhao Zeng
Wenhao Zeng
To enhance the emotion feature and improve the performance of speech emotion recognition, an attention mechanism is employed to recognize the important information in both time and feature dimensions. In the time dimension, multi-heads attention is modified with the last state of the long short-term memory (LSTM)'s output to match the time accumulation characteristic of LSTM. In the feature dimension, scaled dot-product attention is replaced with additive attention that refers to the method of the state update of LSTM to construct multi-heads attention. This means that a nonlinear change replaces the linear mapping in classical multi-heads attention. Experiments on IEMOCAP datasets demonstrate that the attention mechanism could enhance emotional information and improve the performance of speech emotion recognition.

15 September 2023 in IEEE Transactions on Cognitive and Developmental Systems
Verified
Journal Article
Closed access
CPCong Pang
Cong Pang
JFJingjie Fan
Jingjie Fan
QSQifan Shen
Qifan Shen
YXYue Xie
Yue Xie
Chengwei HuangChengwei Huang
Chengwei Huang
BSBjörn W. Schuller
Björn W. Schuller
Both spatial and temporal contextual information are essential for the multichannel speech enhancement (MCSE) task. In this work, we propose a unified MCSE network composed of neural beamforming and a context-focused post-filtering network in order to fully exploit these two types of information. The network is used to estimate the optimum complex ideal ratio masks (cIRM) which can more effectively utilize the phase information in the frequency domain to reconstruct the speech waveform. To assign adaptive weights to channels, we first adopt a dilated convolution-based network which simulates the beamforming on the original multichannel input spectrum as the front-end of the multichannel acoustic model. Furthermore, we propose a post-filtering network that inputs the suggested U-Net’s output to a convolutional long short-term memory (ConvLSTM) layer, as it can properly capture contextual information and spatial correlation information of features. We conduct experiments on the VOICES, CHiME-3 and WMIR datasets, respectively. Experiments show that, in various scenarios, the proposed algorithm shows improvements over the previous state-of-the-art algorithms in terms of PESQ, STOI and SI-SNR.

9 November 2024 in Electronics
Verified
Journal Article
Open access
PDF file
JXJi XI
Ji XI
ZXZhe Xu
Zhe Xu
WZWeiqi Zhang
Weiqi Zhang
Li ZhaoLi Zhao
Li Zhao
YXYue Xie
Yue Xie
To address the performance and computational complexity issues in speech enhancement for hearing aids, a speech enhancement algorithm based on a microphone array and a lightweight two-stage convolutional recurrent network (CRN) is proposed. The algorithm consists of two main modules: a beamforming module and a post-filtering module. The beamforming module utilizes directional features and a complex time-frequency long short-term memory (CFT-LSTM) network to extract local representations and perform spatial filtering. The post-filtering module uses analogous encoding and two symmetric decoding structures, with stacked CFT-LSTM blocks in between. It further reduces residual noise and improves filtering performance by passing spatial information through an inter-channel masking module. Experimental results show that this algorithm outperforms existing methods on the generated hearing aid dataset and the CHIME-3 dataset, with fewer parameters and lower model complexity, making it suitable for hearing aid scenarios with limited computational resources.

30 July 2025 in IEEE Transactions on Audio, Speech and Language Processing
Verified
Journal Article
Closed access
YYYang Yang
Yang Yang
RLRuiyu Liang
Ruiyu Liang
YNYe Ni
Ye Ni
YXYue Xie
Yue Xie
CZCairong Zou
Cairong Zou
BSBjörn W. Schuller
Björn W. Schuller
Accurate evaluation of hearing aid speech quality is crucial for optimizing the auditory experience of hearing-impaired people. Aiming at the shortcomings of existing methods that rely on clean reference signals and do not take into account the effects of differences in Prescription Formula (PF), this paper proposes a non-intrusive speech quality evaluation framework based on speech label assistance, and multi-task learning strategy, termed MTSE-LA. The framework effectively mitigates evaluation bias caused by PF variations and effectively improves the prediction accuracy of speech quality metrics. MTSE-LA consists of three core modules: a feature extraction module, a label classification module, and a score prediction module. The feature extraction module extracts deep frame-level features from speech using a joint Convolutional Neural Network and Bidirectional Long Short-term Memory network (CNN-BiLSTM) model. The label classification module, acting as a pre-trained network, identifies PF labels and embeds them into the extracted frame-level features, which are then fed into the speech quality prediction branch of the multi-task score prediction module. Effective prediction of speech intelligibility is achieved by introducing the output vectors of the modulation filter bank to the speech intelligibility prediction branch to ensure synergy in the multi-task learning process. Moreover, each prediction branch uses the multi-head self-attention mechanism to capture contextual information and model the importance of speech frames. Experimental results demonstrate that MTSE-LA considerably improves the prediction accuracy of the Hearing Aid Speech Quality Index (HASQI) under multiple PF configurations and different degrees of hearing loss conditions. Compared with existing cutting-edge methods, the proposed framework exhibits higher correlation and fitting accuracy, establishing its reliability and superiority in the field of non-intrusive speech quality evaluation for hearing aids.

1 July 2025 in IEICE Transactions on Information and Systems
Verified
Journal Article
Open access
PDF file
RHRong Huang
Rong Huang
ZQZewen Qian
Zewen Qian
HMHao Ma
Hao Ma
ZHZhezhe Han
Zhezhe Han
YXYue Xie
Yue Xie
High-precision sports performance prediction of college students is significant in formulating scientific training mechanisms and reasonable physical fitness courses. Traditional sports performance prediction primarily relies on subjective experience, which suffers from limitations such as individual variations and low reliability. To overcome these shortcomings, an ensemble learning algorithm is proposed in this study. Firstly, a historical dataset for college students is established, including physical characteristics (such as age, height, weight, and lung capacity) and sports performance (such as the 50-meter dash, 1000-meter run, and standing long jump). Then, three forecasting engines including support vector regression, extreme learning machine and decision tree are employed for preliminary prediction based on the preprocessed physical characteristics. Sequentially, three preliminary predictions are combined by the Gaussian process regression in a nonlinear manner to achieve a final probabilistic prediction. By using the data collected from college students, the feasibility of the established ensemble model is evaluated. Practical confirms that the proposed model can fix the performance gap of the individual forecasting engine, effectively improving the prediction accuracy. In addition, the proposed method not only provides point predictions but also generates confidence interval information, which greatly enhances the prediction reliability.