Although automatic speech recognition (ASR) can perform well in common non-overlapping environments, sustaining performance in multi-speaker overlapping speech recognition remains challenging. Recent research revealed that ASR model's encoder captures different levels of information with different layers -- the lower layers tend to have more acoustic information, and the upper layers more linguistic. This inspires us to develop a Sidecar separator to empower a well-trained ASR model for multi-speaker scenarios by separating the mixed speech embedding between two suitable layers. We experimented with a wav2vec 2.0-based ASR model with a Sidecar mounted. By freezing the parameters of the original model and training only the Sidecar (8.7 M, 8.4% of all parameters), the proposed approach outperforms the previous state-of-the-art by a large margin for the 2-speaker mixed LibriMix dataset, reaching a word error rate (WER) of 10.36%; and obtains comparable results (7.56%) for LibriSpeechMix dataset when limited training.
Query-focused meeting summarization (QFMS) aims to generate summaries from meeting transcripts in response to a given query. Previous works typically concatenate the query with meeting transcripts and implicitly model the query relevance only at the token level with attention mechanism. However, due to the dilution of key query-relevant information caused by long meeting transcripts, the original transformer-based model is insufficient to highlight the key parts related to the query. In this paper, we propose a query-aware framework with joint modeling token and utterance based on Query-Utterance Attention. It calculates the utterance-level relevance to the query with a dense retrieval module. Then both token-level query relevance and utterance-level query relevance are combined and incorporated into the generation process with attention mechanism explicitly. We show that the query relevance of different granularities contributes to generating a summary more related to the query. Experimental results on the QMSum dataset show that the proposed model achieves new state-of-the-art performance.
Membership inference (MI) attack is currently the most popular test for measuring privacy leakage in machine learning models. Given a machine learning model, a data point and some auxiliary information, the goal of an MI attack is to determine whether the data point was used to train the model. In this work, we study the reliability of membership inference attacks in practice. Specifically, we show that a model owner can plausibly refute the result of a membership inference test on a data point $x$ by constructing a proof of repudiation that proves that the model was trained without $x$. We design efficient algorithms to construct proofs of repudiation for all data points of the training dataset. Our empirical evaluation demonstrates the practical feasibility of our algorithm by constructing proofs of repudiation for popular machine learning models on MNIST and CIFAR-10. Consequently, our results call for a re-evaluation of the implications of membership inference attacks in practice.
Synthesizing face images from monochrome sketches is one of the most fundamental tasks in the field of image-to-image translation. However, it is still challenging to (1)~make models learn the high-dimensional face features such as geometry and color, and (2)~take into account the characteristics of input sketches. Existing methods often use sketches as indirect inputs (or as auxiliary inputs) to guide the models, resulting in the loss of sketch features or the alteration of geometry information. In this paper, we introduce a Sketch-Guided Latent Diffusion Model (SGLDM), an LDM-based network architect trained on the paired sketch-face dataset. We apply a Multi-Auto-Encoder (AE) to encode the different input sketches from different regions of a face from pixel space to a feature map in latent space, which enables us to reduce the dimension of the sketch input while preserving the geometry-related information of local face details. We build a sketch-face paired dataset based on the existing method that extracts the edge map from an image. We then introduce a Stochastic Region Abstraction (SRA), an approach to augment our dataset to improve the robustness of SGLDM to handle sketch input with arbitrary abstraction. The evaluation study shows that SGLDM can synthesize high-quality face images with different expressions, facial accessories, and hairstyles from various sketches with different abstraction levels.
Internet memes are characterised by the interspersing of text amongst visual elements. State-of-the-art multimodal meme classifiers do not account for the relative positions of these elements across the two modalities, despite the latent meaning associated with where text and visual elements are placed. Against two meme sentiment classification datasets, we systematically show performance gains from incorporating the spatial position of visual objects, faces, and text clusters extracted from memes. In addition, we also present facial embedding as an impactful enhancement to image representation in a multimodal meme classifier. Finally, we show that incorporating this spatial information allows our fully automated approaches to outperform their corresponding baselines that rely on additional human validation of OCR-extracted text.
Within the network analysis field, network immunization refers to the task of protecting a network from some arbitrary diffusion that tries to infect it. In this article, we consider the spread of harmful content in social networks, and we propose CONTAIN, a novel COmmuNiTy-based Algorithm for network ImmuNization. Our solution uses the network information to (1) detect harmful content spreaders, and (2) generate partitions and rank them for immunization using the subgraphs induced by each spreader, i.e., employing CONTAIN. The experimental results obtained on real-world datasets show that CONTAIN outperforms state-of-the-art solutions, i.e., NetShield and SparseShield, by immunizing the network in fewer iterations, thus, converging significantly faster than the state-of-the-art algorithms.
General-purpose embedding is highly desirable for few-shot even zero-shot learning in many application scenarios, including audio tasks. In order to understand representations better, we conducted a thorough error analysis and visualization of HEAR 2021 submission results. Inspired by the analysis, this work experiments with different front-end audio preprocessing methods, including Constant-Q Transform (CQT) and Short-time Fourier transform (STFT), and proposes a Batch Embedding Covariance Regularization (BECR) term to uncover a more holistic simulation of the frequency information received by the human auditory system. We tested the models on the suite of HEAR 2021 tasks, which encompass a broad category of tasks. Preliminary results show (1) the proposed BECR can incur a more dispersed embedding on the test set, (2) BECR improves the PaSST model without extra computation complexity, and (3) STFT preprocessing outperforms CQT in all tasks we tested. Github:https://github.com/ankitshah009/general_audio_embedding_hear_2021
In noisy and reverberant environments, the performance of deep learning-based speech separation methods drops dramatically because previous methods are not designed and optimized for such situations. To address this issue, we propose a multi-stage end-to-end learning method that decouples the difficult speech separation problem in noisy and reverberant environments into three sub-problems: speech denoising, separation, and de-reverberation. The probability and speed of searching for the optimal solution of the speech separation model are improved by reducing the solution space. Moreover, since the channel information of the audio sequence in the time domain is crucial for speech separation, we propose a triple-path structure capable of modeling the channel dimension of audio sequences. Experimental results show that the proposed multi-stage triple-path method can improve the performance of speech separation models at the cost of little model parameter increment.
Decision-makers in the humanitarian sector rely on timely and exact information during crisis events. Knowing how many civilians were injured during an earthquake is vital to allocate aids properly. Information about such victim counts is often only available within full-text event descriptions from newspapers and other reports. Extracting numbers from text is challenging: numbers have different formats and may require numeric reasoning. This renders purely string matching-based approaches insufficient. As a consequence, fine-grained counts of injured, displaced, or abused victims beyond fatalities are often not extracted and remain unseen. We cast victim count extraction as a question answering (QA) task with a regression or classification objective. We compare regex, dependency parsing, semantic role labeling-based approaches, and advanced text-to-text models. Beyond model accuracy, we analyze extraction reliability and robustness which are key for this sensitive task. In particular, we discuss model calibration and investigate few-shot and out-of-distribution performance. Ultimately, we make a comprehensive recommendation on which model to select for different desiderata and data domains. Our work is among the first to apply numeracy-focused large language models in a real-world use case with a positive impact.
Massive antenna arrays form physically large apertures with a beam-focusing capability, leading to outstanding wireless power transfer (WPT) efficiency paired with low radiation levels outside the focusing region. However, leveraging these features requires accurate knowledge of the multipath propagation channel and overcoming the (Rayleigh) fading channel present in typical application scenarios. For that, reciprocity-based beamforming is an optimal solution that estimates the actual channel gains from pilot transmissions on the uplink. But this solution is unsuitable for passive backscatter nodes that are not capable of sending any pilots in the initial access phase. Using measured channel data from an extremely large-scale MIMO (XL-MIMO) testbed, we compare geometry-based planar wavefront and spherical wavefront beamformers with a reciprocity-based beamformer, to address this initial access problem. We also show that we can predict specular multipath components (SMCs) based only on geometric environment information. We demonstrate that a transmit power of 1W is sufficient to transfer more than 1mW of power to a device located at a distance of 12.3m when using a (40x25) array at 3.8GHz. The geometry-based beamformer exploiting predicted SMCs suffers a loss of only 2dB compared with perfect channel state information.