The rise of personal speech data collection has enabled malicious actors to manipulate content using AI-powered tools via speech tampering techniques, such as copy-move forgery, deleting, homologous splicing, and heterologous splicing operations. These made-up things lead to fake voices, cheating, and spreading lies. While deep neural networks (DNNs) show promise for blind speech tampering detection, current methods lack cross-type generalisation due to software-specific limitations and rarely consider the classification of tampering types. To address this issue, this work presents an integrated detection framework (IDF) based on DNNs with three key components: 1) an encoder-decoder detection network generating a preliminary localisation mask (PLM) from Mel spectrogram features (MSFs); 2) an adaptive speech tampering localisation algorithm for refining the PLM; and 3) a dedicated classification network for tampering-type identification. Our IDF is trained using only the basic MSFs, ground-truth masks (GTMs), integer-class labels, binary cross-entropy loss, and sparse categorical cross-entropy loss. Supporting this work, we have produced a comprehensive dataset comprising 55,500 pairs of MSFs and GTMs for authentic and tampered speech samples derived from Chinese and English corpora through systematic tampering simulations. Experimental results demonstrate the consistent performance of our IDF framework across multiple languages and manipulation types. The model achieves average F-scores of 95.32% on Chinese, 96.02% on English, 88.81% on Spanish, and 95.75% on spoofed samples for detection tasks, while maintaining a localisation error below 0.1 seconds.

The overlapping coalition structure generation problem (OCSGP) is a challenging computational problem in multi-agent systems. It focuses on selecting possibly overlapping coalitions from a set of agents to maximize the social welfare of all coalitions while containing all agents. However, in practical applications, coalitions may be formed to selectively respond to tasks from a pool of potential tasks assigned to agents. Consequently, this study considers OCSGP in a task-based setting, where each agent has finite resources and can only respond to tasks of interest, and each coalition can only take on mutually disjoint subsets of tasks. Specifically, we first present a model of the task-based OCSGP and investigate its computational complexity. Our theoretical results demonstrate that this specific OCSGP remains intractable even under restrictive assumptions. Subsequently, we develop a generic evolutionary algorithm framework (EAF) to find an approximately optimal overlapping coalition structure (OCS) in time quartic polynomial in the size of the instance. Particularly, we devise a specific solution-repair based heuristic of cubic time complexity to generate a feasible OCS. Finally, we compare the proposed EAF with a task-oriented heuristic and a hybrid algorithm for OCSGP, and examine its applicability in the pursuit-evasion problem. The experimental results reveal that the proposed EAF exhibits superior performance in finding feasible OCSs and demonstrates flexible adaptability to problem size and resource status.

Digital audio watermarking is a critical technology widely used for copyright protection, content authentication, and broadcast monitoring. However, its robustness is significantly challenged by recapturing and hybrid attacks, which can easily remove watermarks. To address this issue, this work proposes a novel scheme based on the light gradient boosting machine (LightGBM), named LRAW (LightGBM-based Robust Audio Watermarking), which is designed to increase the robustness of audio watermarking against various attacks. Specifically, the scheme begins by analysing coefficients derived from the discrete wavelet transform (DWT), graph-based transform (GBT), and singular value decomposition (SVD). The extracted singular values consistently maintain a stable descending order even under recapturing attacks at a slightly greater distance. Leveraging this stability, the watermark information is implicitly embedded into the audio signal using a quantization rule. To simulate a hybrid attack scenario, a comprehensive feature dataset comprising 396,000 pieces of DWT-GBT-SVD feature data is constructed based on 60 original recordings and 9 types of attack. Furthermore, considering the distinct influences of embedding watermark bits 0 and 1 on the quantization of singular values, the watermark extraction process is formulated as a binary classification problem. LightGBM is trained using Bayesian optimization and the feature dataset to classify the watermark bits accurately. Finally, the complete watermark is recovered using a watermark sequence matching algorithm. Theoretical analysis and experimental results demonstrate that the proposed LRAW scheme outperforms state-of-the-art watermarking methods in robustness against various recapturing and hybrid attacks, even when the distance between the acoustic source and the receiver is considerable.

Advancements in speech synthesis technology bring generated speech closer to natural human voices, but they also introduce a series of potential risks, such as the dissemination of false information and voice impersonation. Therefore, it becomes significant to detect any potential misuse of the released speech content. This letter introduces an active strategy that combines audio watermarking with the HiFi-GAN vocoder to embed an invisible watermark in all synthesized speech for detection purposes. We first pre-train a watermark extraction network as the watermark extractor, and then use the watermark extraction loss and speech quality loss of the extractor to adjust the HiFi-GAN generator to ensure that the watermark can be extracted from the synthesized speech. We evaluate the imperceptibility and robustness of the watermark across various speech synthesis models. The experimental results demonstrate that our method effectively withstands various attacks and exhibits excellent imperceptibility. Moreover, our method is universal and compatible with various vocoder-based speech synthesis models.

Generative steganography has become a popular research topic in the field of generative AI, including generative image and synthetic speech steganography. However, music files have different statistical properties and knowledge representation compared to image and speech files, and the reversible transform between secret message and music is also challenging. Therefore, the existing generative steganographic methods that are effective for image/speech may not be directly effective for music. In this paper, we propose a generative music steganography method, named MIDI-GAN, to generate a secret message as an artificial stego MIDI file using generative adversarial networks (GANs). The created stego MIDI file is small in size, has sweet melodies, and is undetectable to deep learning-based steganalyzers. Unlike the previous generative image/speech steganography, the stego MIDI can also be presented as a sequence of chord numbers, making it difficult for anyone to detect and see grounds for suspicion. Moreover, these chord numbers can be transmitted as any other digital or physical medium to evade detection. Specifically, MIDI-GAN comprises a generator, a discriminator, and an extractor. The generator synthesizes a stego MIDI file from the secret message, while the discriminator ensures that the stego MIDI file approaches the authentic rather than the synthetic MIDI file as much as possible in statistical distribution. The extractor recovers the secret message from the stego MIDI file or chord sequence. Experimental results demonstrate that MIDI-GAN has high concealment and security, as the stego MIDI generated by our method is closely similar to the authentic MIDI files and maintains excellent anti-detection ability against deep learning-based steganalysis.

The multi-objective testing resource allocation problem (MOTRAP) is concerned on how to reasonably plan the testing time of software testers to save the cost and improve the reliability as much as possible. The feasible solution space of a MOTRAP is determined by its variables (i.e., the time invested in each component) and constraints (e.g., the pre-specified reliability, cost, or time). Although a variety of state-of-the-art constrained multi-objective optimisers can be used to find individual solutions in this space, their search remains inefficient and expensive due to the fact that this space is very tiny compared to the large search space. The decision maker may often suffer a prolonged but unsuccessful search that fails to return a feasible solution. In this work, we first formulate a heavily constrained MOTRAP on the basis of an architecture-based model, in which reliability, cost, and time are optimised under the pre-specified multiple constraints on reliability, cost, and time. Then, to estimate the feasible solution space of this specific MOTRAP, we develop theoretical and algorithmic approaches to deduce new tighter lower and upper bounds on variables from constraints. Importantly, our approach can help the decision maker identify whether their constraint settings are practicable, and meanwhile, the derived bounds can just enclose the tiny feasible solution space and help off-the-shelf constrained multi-objective optimisers make the search within the feasible solution space as much as possible. Additionally, to further make good use of these bounds, we propose a generalised bound constraint handling method that can be readily employed by constrained multi-objective optimisers to pull infeasible solutions back into the estimated space with theoretical guarantee. Finally, we evaluate our approach on application and empirical cases. Experimental results reveal that our approach significantly enhances the efficiency, effectiveness, and robustness of off-the-shelf constrained multi-objective optimisers and state-of-the-art bound constraint handling methods at finding high-quality solutions for the decision maker. These improvements may help the decision maker take the stress out of setting constraints and selecting constrained multi-objective optimisers and facilitate the testing planning more efficiently and effectively.

Audio recordings used as evidence have become increasingly important to litigation. Before their admissibility as evidence, an audio forensic expert is often required to help determine whether the submitted audio recordings are altered or authentic. Within this field, the copy-move forgery detection (CMFD), which focuses on finding possible forgeries that are derived from the same audio recording, has been an urgent problem in blind audio forensics. However, most of the existing methods require idealistic pre-segmentation and artificial threshold selection to calculate the similarity between segments, which may result in serious misleading and misjudgment especially on high frequency words. In this work, we present a robust method for detecting and locating an audio copy-move forgery on the basis of constant Q spectral sketches (CQSS) and the integration of a customised genetic algorithm (GA) and support vector machine (SVM). Specifically, the CQSS features are first extracted by averaging the logarithm of the squared-magnitude constant Q transform. Then, the CQSS feature set is automatically optimised by a customised GA combined with SVM to obtain the best feature subset and classification model at the same time. Finally, the integrated method, named CQSS-GA-SVM, is evaluated against the state-of-the-art approaches to blind detection of copy-move forgeries on real-world copy-move datasets with read English and Chinese corpus, respectively. The experimental results demonstrate that the proposed CQSS-GA-SVM exhibits significantly high robustness against post-processing based anti-forensics attacks and adaptability to the changes of the duplicated segment duration, the training set size, the recording length, and the forgery type, which may be beneficial to improving the work efficiency of audio forensic experts.

Large-scale cropping (LSC) is one of the mostly-used operations in desynchronization attacks and can easily destroy the watermark information by deleting continuous audio slices from the watermarked audio. In this work, we propose a spread spectrum (SS) based audio watermarking scheme to resist against LSC attacks more robustly from both theoretical and empirical perspectives. Specifically, we first perform discrete wavelet transform (DWT), graph-based transform (GBT), and singular value decomposition (SVD) on the host audio signal to produce transform coefficients. Next, we embed the chaotic encrypted watermark into DWT-GBT-SVD coefficients by the SS technique. Then, we combine m-sequences with the encrypted watermark to generate the watermarking key, which can theoretically guarantee the self-restoration of the cropped watermark based on the periodicity of m-sequences. Additionally, we develop an effective sliding window (SW) strategy to extract the fragmentary watermark slices from DWT-GBT-SVD coefficients and restore the integral watermark by the watermarking key. Finally, the proposed audio watermarking scheme, named m-SW-LSC, is compared with the state-of-the-art audio watermarking methods on audio signals with different genres and lengths under various attacks with different ratios. Experimental results demonstrate that our m-SW-LSC has a superior performance in restoring the complete watermark and exhibits significantly high robustness against LSC attacks.

The multi-objective testing resource allocation problem (MOTRAP) aims at seeking a good trade-off between system reliability, testing cost, and testing time, which is of significant importance to facilitate the testing planning. Yet most studies focus on the time constraint but rarely consider the practical reliability requirement. In this work, we address MOTRAP on an architecture-based model (ABM) with the personalized preference over reliability. More specifically, we first present a reliability-constrained MOTRAP model on the basis of ABM and illustrate how to use this model for real-world systems. Then, to leverage the problem's knowledge, we develop new lower and upper bounds on testing time invested in different components from both theoretical and algorithmic perspectives on the basis of the Lagrange multiplier and half-interval search. Importantly, these new derived bounds have strong implications due to the fact that they can be easily employed by optimizers as the limits of variables to prune the search space to the region of interests of the decision maker and locate feasible solutions with the expected reliability. Finally, we evaluate the proposed bounds in popular multi-objective optimizers for MOTRAP on application and empirical cases. Experimental results demonstrate that our new bounds practically improve the search performance of optimizers, and decision makers can easily combine these new bounds with off-the-shelf optimizers to find higher-quality solutions that they are interested in, which greatly soothes away stress on optimizer and solution selections of decision makers.

The multiobjective testing resource allocation problem (MOTRAP) is how to efficiently allocate the finite testing time to various modules, with the aim of optimizing system reliability, testing cost, and testing time simultaneously. To deal with this problem, a common approach is to use multiobjective evolutionary algorithms (MOEAs) to seek a set of tradeoff solutions between the three objectives. However, such a tradeoff set may contain a substantial proportion of solutions with very low reliability level, which consume lots of computational resources but may be valueless to the software project manager. In this article, a MOTRAP model with a prespecified reliability is first proposed. Then, new lower bounds on the testing time invested in different modules are theoretically deduced from the necessary condition for the achievement of the given reliability, based on which an exact algorithm for determining the new lower bounds is presented. Moreover, several enhanced constraint-handling techniques (ECHTs) derived from the new bounds are successively developed to be combined with MOEAs to correct and reduce the constraint violation. Finally, the proposed ECHTs are evaluated in comparison with various state-of-the-art constraint-solving approaches. The comparative results demonstrate that the proposed ECHTs can work well with MOEAs, make the search focus on the feasible region of the prespecified reliability, and provide the software project manager with better and more diverse, satisfactory choices in test planning.

Coalition formation has been a fundamental form of resource cooperation for achieving joint goals in multiagent systems. Most existing studies still focus on the traditional assumption that an agent has to contribute its resources to all the goals, even if the agent is not interested in the goal at all. In this article, a natural extension of the traditional coalitional resource games (CRGs) is studied from both theoretical and empirical perspectives, in which each agent has uncompromising, personalized preferences over goals. Specifically, a new CRGs model with agents’ strict preferences for goals is presented, in which an agent is willing to contribute its resources only to the goals that are in its own interest set. The computational complexity of the basic decision problems surrounding the successful coalition is reinvestigated. The results suggest that these problems in such a strict preference way are complex and intractable. To find the largest successful coalition for possible computation reduction or potential parallel processing, a flow-network–based exhaust algorithm, called FNetEA, is proposed to achieve the optimal solution. Then, to solve the problem more efficiently, a hybrid algorithm, named 2D-HA, is developed to find the approximately optimal solution on the basis of genetic algorithm, two-dimensional (2D) solution representation, and a heuristic for solution repairs. Through extensive experiments, the 2D-HA algorithm exhibits the prominent ability to provide reassurances that the optimal solution could be found within a reasonable period of time, even in a super-large-scale space.

Overlapping coalition formation (OCF), which provides a natural framework for modeling scenarios where each agent can join and allocate their resources to several completely different coalitions at the same time, has become a very active topic in multiagent systems. For OCF in resource-constrained and subadditive task oriented domains, an agent may not possess sufficient resources to meet the needs of multiple coalitions simultaneously. As a result, there may exist many potential resource conflicts among the rival overlapping coalitions. To tackle such situations, we first present a natural variation of the traditional OCF model and analyze the size of the solution space and the computational complexity of the overlapping coalition structure generation (OCSG) problem. Next, we develop a generic task-oriented heuristic (TOH) for individual repairs that can be used in binary meta-heuristic algorithms to generate overlapping coalitions in a parallel manner. Moreover, we show how the proposed TOH repairs a 2-D individual to resolve resource conflicts and discuss several basic properties. Finally, to evaluate the effectiveness of TOH, we compare it with the existing agent-oriented heuristic for the OCSG problem. The empirical results demonstrate that TOH is of high efficiency and effectiveness in harsh environments with fierce competition over scarce resources.

In spread spectrum (SS) based robust audio watermarking, the scaling parameter is an important factor for balancing between robustness and imperceptibility. There have been intense studies of the embedded parameter optimization in light of the signal-to-noise ratio (SNR), but little attention has been given to the constrained SNR. Moreover, traditional population-based stochastic search algorithms for optimizing the embedded parameter significantly increase the computation pressure of the corresponding audio watermarking schemes. This paper comprehensively investigates the effect of the constrained SNR on the optimization of the scaling parameter, from both model and algorithmic perspectives. Specifically, the empirical relationship between the scaling parameter, robustness, and imperceptibility is first analyzed in detail. Next, an SNR-constrained optimization model is presented. Then, to solve the proposed model and find the current optimal scaling parameter for watermark embedding, a binary search algorithm and a heuristic search (HS) algorithm are, respectively, developed. Finally, we embed the proposed model and heuristics in the SS-based audio watermarking scheme and compare the integrated technique (called SS-SNR-HS) with the existing similar schemes. The experimental results demonstrate that SS-SNR-HS not only is computationally simple, but also achieves better balance between imperceptibility and robustness and, thus, seems promising in copyright protection of online digital audio.

In software testing, optimal testing resource allocation problems (OTRAPs) are important when seeking a good tradeoff between reliability, cost, and time with limited resources. There have been intensive studies of OTRAPs using multiobjective evolutionary algorithms (MOEAs), but little attention has been paid to the constraint handling. This paper comprehensively investigates the effect of the constraint handling on the performance of nondominated sorting genetic algorithm II (NSGA-II) for solving OTRAPs, from both theoretical and empirical perspectives. The heuristics for individual repairs are first proposed to handle constraint violations in NSGA-II, based on which several properties are derived. Additionally, theZ-score based Euclidean distance is adopted to estimate the difference between solutions. Finally, the above methods are evaluated and the experiments show several results. 1) The developed heuristics for constraint handling are better than the Existing Strategy in terms of the capacity and coverage values. 2) TheZ-score operation obtains better diversity values and reduces repeated solutions. 3) The modified NSGA-II for OTRAPs (called NSGA-II-TRA) performs significantly better than the existing MOEAs in terms of capacity and coverage values, which suggests that NSGA-II-TRA could obtain more and higher quality testing-time-allocation schemes, especially for large, complex datasets. 4) NSGA-II-TRA is robust according to the sensitivity analysis results.