Hengshun Zhou
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Hengshun Zhou has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
- Jun Du
University of Science and Technology of China, Hefei, Anhui, China
4 shared publications - Hang Chen
University of Science and Technology of China, Hefei, Anhui, China
2 shared publications - Ke-Wei Li
NERC-SLIP, USTC, Hefei, Anhui, China
2 shared publications - Qing Wang
University of Science and Technology of China, Hefei, China
2 shared publications - Shi Cheng
FlyTek, Hefei, Anhui, China
2 shared publications - Chen-Yue Zhang
FlyTek, Hefei, Anhui, China
1 shared publication
Latest Publications
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 13148-13155
MISP-QEKS: A Large-Scale Dataset with Multimodal Cues for Query-by-Example Keyword Spotting
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 13148-13155
Open AccessMISP-QEKS: A Large-Scale Dataset with Multimodal Cues for Query-by-Example Keyword Spotting
- 13 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 35 (7) , 7308-7320
Lightweight Audio-Visual Wake Word Spotting With Diverse Acoustic Knowledge Distillation
Audio-Visual Wake Word Spotting (AVWWS) aims to accurately detect user-defined keywords by leveraging the complementary nature of different modalities in challenging acoustic environments. However, two primary challenges hinder the application of AVWWS models in real-world scenarios: increased model parameters involving the video modality and the scarcity of paired audio-visual data. To address these issues, we propose a novel diverse acoustic knowledge distillation (DAKD) framework, which utilizes easily accessible single-modality audio data to train two teacher models and employs cross-modal knowledge distillation to transfer the generalization and de-noising capabilities of the teachers to the audio-visual student model. This approach mitigates the overfitting risk associated with large parameter counts and limited data. The DAKD framework consists of an audio-visual student model based on the lightweight multi-scale temporal-spatial attention (LMTSA) architecture, a multi-conditional teacher (MCT) model, and a de-noising teacher (DNT) model. The LMTSA model integrates compact 3D and 2D blocks based on the ResNet architecture through a simple attention module and accepts multi-scale supervision from word-level and phone-level labels, achieving joint temporal-spatial modeling with minimal parameter usage. The MCT and DNT models were trained using extensive real or simulated far-field speech and paired near-field and far-field speech, respectively, to generalize unseen acoustic environments and de-noising capabilities to the audio-visual student model. The effectiveness of our proposed DAKD framework is validated through comprehensive experiments on the MISP2021 and the updated MISP2021 Eval Hard datasets, establishing new benchmarks with fewer parameters. Our code will be available at https://github.com/wikkk-tp/AVWWS_DAKD.
- 13 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 35 (7) , 7308-7320
Lightweight Audio-Visual Wake Word Spotting With Diverse Acoustic Knowledge Distillation
Audio-Visual Wake Word Spotting (AVWWS) aims to accurately detect user-defined keywords by leveraging the complementary nature of different modalities in challenging acoustic environments. However, two primary challenges hinder the application of AVWWS models in real-world scenarios: increased model parameters involving the video modality and the scarcity of paired audio-visual data. To address these issues, we propose a novel diverse acoustic knowledge distillation (DAKD) framework, which utilizes easily accessible single-modality audio data to train two teacher models and employs cross-modal knowledge distillation to transfer the generalization and de-noising capabilities of the teachers to the audio-visual student model. This approach mitigates the overfitting risk associated with large parameter counts and limited data. The DAKD framework consists of an audio-visual student model based on the lightweight multi-scale temporal-spatial attention (LMTSA) architecture, a multi-conditional teacher (MCT) model, and a de-noising teacher (DNT) model. The LMTSA model integrates compact 3D and 2D blocks based on the ResNet architecture through a simple attention module and accepts multi-scale supervision from word-level and phone-level labels, achieving joint temporal-spatial modeling with minimal parameter usage. The MCT and DNT models were trained using extensive real or simulated far-field speech and paired near-field and far-field speech, respectively, to generalize unseen acoustic environments and de-noising capabilities to the audio-visual student model. The effectiveness of our proposed DAKD framework is validated through comprehensive experiments on the MISP2021 and the updated MISP2021 Eval Hard datasets, establishing new benchmarks with fewer parameters. Our code will be available at https://github.com/wikkk-tp/AVWWS_DAKD.
- Open Access
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 9531-9535
Hierarchical Audio-Visual Information Fusion with Multi-label Joint Decoding for MER 2023
In this paper, we propose a novel framework for recognizing both discrete and dimensional emotions. In our framework, deep features extracted from foundation models are used as robust acoustic and visual representations of raw video. Three different structures based on attention-guided feature gathering (AFG) are designed for deep feature fusion. Then, we introduce a joint decoding structure for emotion classification and valence regression in the decoding stage. A multi-task loss based on uncertainty is also designed to optimize the whole process. Finally, by combining three different structures on the posterior probability level, we obtain the final predictions of discrete and dimensional emotions. When tested on the dataset of multimodal emotion recognition challenge (MER 2023), the proposed framework yields consistent improvements in both emotion classification and valence regression. Our final system achieves state-of-the-art performance and ranks third on the leaderboard on MER-MULTI sub-challenge.
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 9531-9535
Open AccessHierarchical Audio-Visual Information Fusion with Multi-label Joint Decoding for MER 2023
In this paper, we propose a novel framework for recognizing both discrete and dimensional emotions. In our framework, deep features extracted from foundation models are used as robust acoustic and visual representations of raw video. Three different structures based on attention-guided feature gathering (AFG) are designed for deep feature fusion. Then, we introduce a joint decoding structure for emotion classification and valence regression in the decoding stage. A multi-task loss based on uncertainty is also designed to optimize the whole process. Finally, by combining three different structures on the posterior probability level, we obtain the final predictions of discrete and dimensional emotions. When tested on the dataset of multimodal emotion recognition challenge (MER 2023), the proposed framework yields consistent improvements in both emotion classification and valence regression. Our final system achieves state-of-the-art performance and ranks third on the leaderboard on MER-MULTI sub-challenge.
- 14 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 29 (23299290) , 2617-2629
Information Fusion in Attention Networks Using Adaptive and Multi-Level Factorized Bilinear Pooling for Audio-Visual Emotion Recognition
Multimodal emotion recognition is a challenging task in emotion computing as it is quite difficult to extract discriminative features to identify the subtle differences in human emotions with abstract concept and multiple expressions. Moreover, how to fully utilize both audio and visual information is still an open problem. In this paper, we propose a novel multimodal fusion attention network for audio-visual emotion recognition based on adaptive and multi-level factorized bilinear pooling (FBP). First, for the audio stream, a fully convolutional network (FCN) equipped with 1-D attention mechanism and local response normalization is designed for speech emotion recognition. Next, a global FBP (G-FBP) approach is presented to perform audio-visual information fusion by integrating self-attention based video stream with the proposed audio stream. To improve G-FBP, an adaptive strategy (AG-FBP) to dynamically calculate the fusion weight of two modalities is devised based on the emotion-related representation vectors from the attention mechanism of respective modalities. Finally, to fully utilize the local emotion information, adaptive and multi-level FBP (AM-FBP) is introduced by combining both global-trunk and intra-trunk data in one recording on top of AG-FBP. Tested on the IEMOCAP corpus for speech emotion recognition with only audio stream, the new FCN method outperforms the state-of-the-art results with an accuracy of 71.40%. Moreover, validated on the AFEW database of EmotiW2019 sub-challenge and the IEMOCAP corpus for audio-visual emotion recognition, the proposed AM-FBP approach achieves the best accuracy of 63.09% and 75.49% respectively on the test set.
- 14 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 29 (23299290) , 2617-2629
Information Fusion in Attention Networks Using Adaptive and Multi-Level Factorized Bilinear Pooling for Audio-Visual Emotion Recognition
Multimodal emotion recognition is a challenging task in emotion computing as it is quite difficult to extract discriminative features to identify the subtle differences in human emotions with abstract concept and multiple expressions. Moreover, how to fully utilize both audio and visual information is still an open problem. In this paper, we propose a novel multimodal fusion attention network for audio-visual emotion recognition based on adaptive and multi-level factorized bilinear pooling (FBP). First, for the audio stream, a fully convolutional network (FCN) equipped with 1-D attention mechanism and local response normalization is designed for speech emotion recognition. Next, a global FBP (G-FBP) approach is presented to perform audio-visual information fusion by integrating self-attention based video stream with the proposed audio stream. To improve G-FBP, an adaptive strategy (AG-FBP) to dynamically calculate the fusion weight of two modalities is devised based on the emotion-related representation vectors from the attention mechanism of respective modalities. Finally, to fully utilize the local emotion information, adaptive and multi-level FBP (AM-FBP) is introduced by combining both global-trunk and intra-trunk data in one recording on top of AG-FBP. Tested on the IEMOCAP corpus for speech emotion recognition with only audio stream, the new FCN method outperforms the state-of-the-art results with an accuracy of 71.40%. Moreover, validated on the AFEW database of EmotiW2019 sub-challenge and the IEMOCAP corpus for audio-visual emotion recognition, the proposed AM-FBP approach achieves the best accuracy of 63.09% and 75.49% respectively on the test set.