Search

The Multi-modal Information based Speech Processing (MISP) challenge aims to extend the application of signal processing technology in specific scenarios by promoting the research into wake-up words, speaker diarization, speech recognition, and other technologies. The MISP2022 challenge has two tracks: 1) audio-visual speaker diarization (AVSD), aiming to solve ``who spoken when'' using both audio and visual data; 2) a novel audio-visual diarization and recognition (AVDR) task that focuses on addressing ``who spoken what when'' with audio-visual speaker diarization results. Both tracks focus on the Chinese language, and use far-field audio and video in real home-tv scenarios: 2-6 people communicating each other with TV noise in the background. This paper introduces the dataset, track settings, and baselines of the MISP2022 challenge. Our analyses of experiments and examples indicate the good performance of AVDR baseline system, and the potential difficulties in this challenge due to, e.g., the far-field video quality, the presence of TV noise in the background, and the indistinguishable speakers., 5 pages, 4 figures, to be published in ICASSP2023

Cobalt-based superalloys have good service performance in environments where nickel-based superalloys cannot work. A number of the studies focused on mechanical properties, microstructure evolution and hot deformation characteristics, but studies of meta-dynamic recrystallization (MDRX) behavior remain lacking. The MDRX behavior of GH5188 superalloy was investigated using two-stage hot compression tests. The kinetic model and grain size model were built to better describe the MDRX process. The good agreement between the predicted values and experimental values was verified. The effects of holding time, temperature, strain rate and intervals time on MDRX microstructure evolution were analyzed and discussed. The results show that by descending order of correlation was: intervals time, holding time, temperature and strain rate. The MDRX fraction and grain size increase with the increase of holding time, temperature, strain rate and intervals time. At low temperatures, MDRX grains grew up in the form of bowed grain boundaries. At high temperatures, the growth of MDRX grains was often accompanied by the generation of twinning bands. Twins were formed in the process of grain growth.

Personal sound zone systems require to know the acoustic transfer functions (ATFs) between the loudspeakers and matching microphones, which in general have to be measured in advance. As the measurement is affected by such factors as sensors' position mismatches and environmental changes, deviation of the measured ATF-s from the true ones is unavoidable and such perturbation may lead to significant performance degradation in sound zone production. In this work, a method is developed for robust acoustic contrast control, which simultaneously considers the norm-bounded uncertainties and a positive semidefinite constraint to ensure the optimization problem to have a solution. Since the problem is non-convex, the polynomial time difference-of-convex functions (POTDC) algorithm is adopted to identify the optimal solution. Simulations are performed to validate the effectiveness of the proposed method and it demonstrates a performance superior to other existing robust control methods.

As far as Scene Graph Generation (SGG), coarse and fine predicates mix in the dataset due to the crowd-sourced labeling, and the long-tail problem is also pronounced. Given this tricky situation, many existing SGG methods treat the predicates equally and learn the model under the supervision of mixed-granularity predicates in one stage, leading to relatively coarse predictions. In order to alleviate the negative impact of the suboptimum mixed-granularity annotation and long-tail effect problems, this paper proposes a novel Hierarchical Memory Learning (HML) framework to learn the model from simple to complex, which is similar to the human beings' hierarchical memory learning process. After the autonomous partition of coarse and fine predicates, the model is first trained on the coarse predicates and then learns the fine predicates. In order to realize this hierarchical learning pattern, this paper, for the first time, formulates the HML framework using the new Concept Reconstruction (CR) and Model Reconstruction (MR) constraints. It is worth noticing that the HML framework can be taken as one general optimization strategy to improve various SGG models, and significant improvement can be achieved on the SGG benchmark (i.e., Visual Genome).

Nonnegative Tucker Factorization (NTF) minimizes the euclidean distance or Kullback-Leibler divergence between the original data and its low-rank approximation which often suffers from grossly corruptions or outliers and the neglect of manifold structures of data. In particular, NTF suffers from rotational ambiguity, whose solutions with and without rotation transformations are equally in the sense of yielding the maximum likelihood. In this paper, we propose three Robust Manifold NTF algorithms to handle outliers by incorporating structural knowledge about the outliers. They first applies a half-quadratic optimization algorithm to transform the problem into a general weighted NTF where the weights are influenced by the outliers. Then, we introduce the correntropy induced metric, Huber function and Cauchy function for weights respectively, to handle the outliers. Finally, we introduce a manifold regularization to overcome the rotational ambiguity of NTF. We have compared the proposed method with a number of representative references covering major branches of NTF on a variety of real-world image databases. Experimental results illustrate the effectiveness of the proposed method under two evaluation metrics (accuracy and nmi).

The development of scene text recognition (STR) in the era of deep learning has been mainly focused on novel architectures of STR models. However, training protocol (i.e., settings of the hyper-parameters involved in the training of STR models), which plays an equally important role in successfully training a good STR model, is under-explored for scene text recognition. In this work, we attempt to improve the accuracy of existing STR models by searching for optimal training protocol. Specifically, we develop a training protocol search algorithm, based on a newly designed search space and an efficient search algorithm using evolutionary optimization and proxy tasks. Experimental results show that our searched training protocol can improve the recognition accuracy of mainstream STR models by 2.7%~3.9%. In particular, with the searched training protocol, TRBA-Net achieves 2.1% higher accuracy than the state-of-the-art STR model (i.e., EFIFSTR), while the inference speed is 2.3x and 3.7x faster on CPU and GPU respectively. Extensive experiments are conducted to demonstrate the effectiveness of the proposed method and the generalization ability of the training protocol found by our search method. Code is available at https://github.com/VDIGPKU/STR_TPSearch.

Malicious applications of deepfakes (i.e., technologies generating target facial attributes or entire faces from facial images) have posed a huge threat to individuals' reputation and security. To mitigate these threats, recent studies have proposed adversarial watermarks to combat deepfake models, leading them to generate distorted outputs. Despite achieving impressive results, these adversarial watermarks have low image-level and model-level transferability, meaning that they can protect only one facial image from one specific deepfake model. To address these issues, we propose a novel solution that can generate a Cross-Model Universal Adversarial Watermark (CMUA-Watermark), protecting a large number of facial images from multiple deepfake models. Specifically, we begin by proposing a cross-model universal attack pipeline that attacks multiple deepfake models iteratively. Then, we design a two-level perturbation fusion strategy to alleviate the conflict between the adversarial watermarks generated by different facial images and models. Moreover, we address the key problem in cross-model optimization with a heuristic approach to automatically find the suitable attack step sizes for different models, further weakening the model-level conflict. Finally, we introduce a more reasonable and comprehensive evaluation method to fully test the proposed method and compare it with existing ones. Extensive experimental results demonstrate that the proposed CMUA-Watermark can effectively distort the fake facial images generated by multiple deepfake models while achieving a better performance than existing methods., Comment: 9 pages, 7 figures, Thirty-Sixth AAAI Conference on Artificial Intelligence, AAAI22

Coupling mechanism is the key link of energy transfer in wireless power transfer system, and the design of its parameters will affect the system’s efficiency. In this paper, by analyzing the efficiency characteristics of the system under LCC/S topology, mutual inductance is taken as the index of coupling mechanism design. Taking the double-D-shaped coupling coil as the research object, the long primary side and short secondary side coil model suitable for moving conditions is established. The influence of the length / width, wire diameter and line spacing of the secondary side coil on mutual inductance is analyzed, and the laying mode of the core is designed. According to the analysis of the simulation results, the design scheme of the coupling mechanism is determined: the length of the secondary side coil is 850mm and the width is 700mm; the primary side core is laid in the middle of the coil; the loss density of the secondary side core is reduced by increasing the thickness of the core partially. The relevant conclusions are verified by experiments.

Recently, phase processing is attracting increasinginterest in speech enhancement community. Some researchersintegrate phase estimations module into speech enhancementmodels by using complex-valued short-time Fourier transform(STFT) spectrogram based training targets, e.g. Complex RatioMask (cRM) [1]. However, masking on spectrogram would violentits consistency constraints. In this work, we prove that theinconsistent problem enlarges the solution space of the speechenhancement model and causes unintended artifacts. ConsistencySpectrogram Masking (CSM) is proposed to estimate the complexspectrogram of a signal with the consistency constraint in asimple but not trivial way. The experiments comparing ourCSM based end-to-end model with other methods are conductedto confirm that the CSM accelerate the model training andhave significant improvements in speech quality. From ourexperimental results, we assured that our method could enha

Proceedings of the ICA 2019 and EAA Euroregio : 23rd International Congress on Acoustics, integrating 4th EAA Euroregio 2019 : 9-13 September 2019, Aachen, Germany / proceedings editor: Martin Ochmann, Michael Vorländer, Janina Fels 23rd International Congress on Acoustics, integrating 4th EAA Euroregio 2019, ICA 2019, Aachen, Germany, 9 Sep 2019 - 13 Sep 2019; Aachen (2019)., Published by Aachen

Proceedings of the ICA 2019 and EAA Euroregio : 23rd International Congress on Acoustics, integrating 4th EAA Euroregio 2019 : 9-13 September 2019, Aachen, Germany / proceedings editor: Martin Ochmann, Michael Vorländer, Janina Fels 23rd International Congress on Acoustics, integrating 4th EAA Euroregio 2019, ICA 2019, Aachen, Germany, 9 Sep 2019 - 13 Sep 2019; Aachen (2019)., Published by Aachen

Proceedings of the ICA 2019 and EAA Euroregio : 23rd International Congress on Acoustics, integrating 4th EAA Euroregio 2019 : 9-13 September 2019, Aachen, Germany / proceedings editor: Martin Ochmann, Michael Vorländer, Janina Fels 23rd International Congress on Acoustics, integrating 4th EAA Euroregio 2019, ICA 2019, Aachen, Germany, 9 Sep 2019 - 13 Sep 2019; Aachen (2019)., Published by Aachen

Proceedings of the ICA 2019 and EAA Euroregio : 23rd International Congress on Acoustics, integrating 4th EAA Euroregio 2019 : 9-13 September 2019, Aachen, Germany / proceedings editor: Martin Ochmann, Michael Vorländer, Janina Fels 23rd International Congress on Acoustics, integrating 4th EAA Euroregio 2019, ICA 2019, Aachen, Germany, 9 Sep 2019 - 13 Sep 2019; Aachen (2019)., Published by Aachen

Users of Fused Filament Fabrication (FFF) can choose from a wide variety of new materials as filament producers continue to introduce new polymer and polymer composite filament into the marketplace. This paper describes a low-cost device capable of measuring the rheological properties of off-the-shelf polymer filament. In this approach, measurements are taken during filament extrusion which are combined with a pressure drop model based on simple shear flow within the FFF nozzle to perform inverse analysis that computes parameters for the power law generalized Newtonian fluid (GNF) model. The applicability of our FFF-filament rheometer is demonstrated with four commercially available polymer filaments by comparing the results to those obtained from a commercial rotational rheometer. A filament characterization approach similar to Melt Flow Index (MFI) is also proposed to assess the extrusion characteristics of materials specific to FFF.

We show that an end-to-end deep learning approach can be used to recognize either English or Mandarin Chinese speech--two vastly different languages. Because it replaces entire pipelines of hand-engineered components with neural networks, end-to-end learning allows us to handle a diverse variety of speech including noisy environments, accents and different languages. Key to our approach is our application of HPC techniques, resulting in a 7x speedup over our previous system. Because of this efficiency, experiments that previously took weeks now run in days. This enables us to iterate more quickly to identify superior architectures and algorithms. As a result, in several cases, our system is competitive with the transcription of human workers when benchmarked on standard datasets. Finally, using a technique called Batch Dispatch with GPUs in the data center, we show that our system can be inexpensively deployed in an online setting, delivering low latency when serving users at scale.

Speech enhancement is a classical problem in signal processing, yet still largely unsolved. Two of the conventional approaches for solving this problem are linear filtering, like the classical Wiener filter, and subspace methods. These approaches have traditionally been treated as different classes of methods and have been introduced in somewhat different contexts. Linear filtering methods originate in stochastic processes, while subspace methods have largely been based on developments in numerical linear algebra and matrix approximation theory.This book bridges the gap between these two classes of methods by showing how the ideas behind subspace methods can be incorporated into traditional linear filtering. In the context of subspace methods, the enhancement problem can then be seen as a classical linear filter design problem. This means that various solutions can more easily be compared and their performance bounded and assessed in terms of noise reduction and speech distortion. The book shows how various filter designs can be obtained in this framework, including the maximum SNR, Wiener, LCMV, and MVDR filters, and how these can be applied in various contexts, like in single-channel and multichannel speech enhancement, and in both the time and frequency domains.

In this paper, the requirements of survivability of MultiHop Wireless Network (MWNeT) was analyzed firstly, then we classify the impacts that affect survivability into three categories: dynamic topology, faults and attacks, and the impacts of these factors are analyzed individually. A simulation environment for the MWNeT towards survivability is designed and implemented as well. Experiments that under the requirements and the impacts we declared are done based on this environment.