PyTorchKR Papers

최신 논문

총 361편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1910.12459

A Bin Encoding Training of a Spiking Neural Network-based Voice Activity Detection

Advances of deep learning for Artificial Neural Networks(ANNs) have led to significant improvements in the performance of digital signal processing systems implemented on digital chips. Although recent progress in low-power chips is remarkable, neuromorphic chips that run Spiking Neural Networks (SNNs) based applications offer an even lower power consumption, as a consequence of the ensuing sparse spike-based coding scheme. In this work, we develop a SNN-based Voice Activity Detection (VAD) system that belongs to the building blocks of any audio and speech processing system. We propose to use the bin encoding, a novel method to convert log mel filterbank bins of single-time frames into spike patterns. We integrate the proposed scheme in a bilayer spiking architecture which was evaluated on the QUT-NOISE-TIMIT corpus. Our approach shows that SNNs enable an ultra low-power implementation of a VAD classifier that consumes only 3.8$\mu$W, while achieving state-of-the-art performance.

arXiv:1906.10956

An Acoustic Emission Activity Detection Method based on Short-Term Waveform Features: Application to Metallic Components under Uniaxial Tensile Test

The Acoustic Emission (AE) phenomenon has been used as a powerful tool with the purpose to either detect, locate or assess damage for a wide range of applications. Derived from its monitoring, one major current challenge on the analysis of the acquired signal is the proper identification and separation of each AE event. Current advanced methods for detecting events are primarily focused on identifying with high accuracy the beginning of the AE wave; however, the detection of the conclusion has been disregarded in the literature. For an automatic continuous detection of events within a data stream, this lack of accuracy for the conclusion of the events generates errors in two critical aspects. In one hand, it deteriorates the accuracy of the measurement of the events duration, truncating the span of the event (undesirable in evaluation applications), and in the other hand, it causes false detections. In this work, an accurate and computationally efficient AE activity detector is presented, using a framework inspired by the area of speech processing, and which provides the required indicators to accurately detect the onset and the end of an AE event. This is achieved by means of a threshold approach that instead of directly operates with the transduced voltage signal it does so over the Short-Term Energy and the Short-Term Zero-Crossing Rate measures of the signal. The STE-ZCR method is developed for an application related to the continuous monitoring of a single AE channel derived from the characterization of metallic components by means of a uniaxial tensile test. Additionally, two experimental test-benches are implemented with the aim to quantify the accuracy and the quality of event detection of the presented method. Finally, the obtained results are compared with four different techniques, representing the current state of the art related to AE activity detection.

arXiv:1809.10450

Sequence Block based Compressed Sensing Multiuser Detection for 5G

Compressed sensing based multiuser detection (CSMUD) is a promising candidate to cope with the massive connectivity requirements of the massive machine type communication (mMTC) in the fifth generation (5G) wireless communication system. It facilitates grant-free non-orthogonal code division multiple access (CDMA) to accommodate massive number of IoT devices. In non-orthogonal CDMA, the users are assigned with non-orthogonal sequences which serve as their signatures. However, the activity detection which is based on the correlation between the spreading sequences, degrades with increase in the number of users, especially in the lower SNR region. In this paper, to improve the performance of the CSMUD, we propose a sequence block based CSMUD, in which block of sequences is used as signature of the user instead of single sequence. A sequence block based group orthogonal matching pursuit algorithm is proposed to jointly detect the activity and data. The proposed scheme reduces the detection error rate (DER) by a magnitude of two at SNR = 10 dB in a system where the number of users are three times more than the number of available resources with each user having activity probability of 0.1. The DER of the proposed scheme is below 10^-2 even at activity probability of 0.16 for sequence length of 20 and overloading factor of 300%. Furthermore, at SNR = 10 dB, the DER of the proposed scheme outperforms the conventional scheme by a magnitude of one for a system with overloading factor of 500%.

arXiv:1802.09005

Frequency domain TRINICON-based blind source separation method with multi-source activity detection for sparsely mixed signals

The TRINICON ('Triple-N ICA for convolutive mixtures') framework is an effective blind signal separation (BSS) method for separating sound sources from convolutive mixtures. It makes full use of the non-whiteness, non-stationarity and non-Gaussianity properties of the source signals and can be implemented either in time domain or in frequency domain, avoiding the notorious internal permutation problem. It usually has best performance when the sources are continuously mixed. In this paper, the offline dual-channel frequency domain TRINICON implementation for sparsely mixed signals is investigated, and a multi-source activity detection is proposed to locate the active period of each source, based on which the filter updating strategy is regularized to improve the separation performance. The objective metric provided by the BSSEVAL toolkit is utilized to evaluate the performance of the proposed scheme.

arXiv:2012.00914

MEVA: A Large-Scale Multiview, Multimodal Video Dataset for Activity Detection

We present the Multiview Extended Video with Activities (MEVA) dataset, a new and very-large-scale dataset for human activity recognition. Existing security datasets either focus on activity counts by aggregating public video disseminated due to its content, which typically excludes same-scene background video, or they achieve persistence by observing public areas and thus cannot control for activity content. Our dataset is over 9300 hours of untrimmed, continuous video, scripted to include diverse, simultaneous activities, along with spontaneous background activity. We have annotated 144 hours for 37 activity types, marking bounding boxes of actors and props. Our collection observed approximately 100 actors performing scripted scenarios and spontaneous background activity over a three-week period at an access-controlled venue, collecting in multiple modalities with overlapping and non-overlapping indoor and outdoor viewpoints. The resulting data includes video from 38 RGB and thermal IR cameras, 42 hours of UAV footage, as well as GPS locations for the actors. 122 hours of annotation are sequestered in support of the NIST Activity in Extended Video (ActEV) challenge; the other 22 hours of annotation and the corresponding video are available on our website, along with an additional 306 hours of ground camera data, 4.6 hours of UAV data, and 9.6 hours of GPS logs. Additional derived data includes camera models geo-registering the outdoor cameras and a dense 3D point cloud model of the outdoor scene. The data was collected with IRB oversight and approval and released under a CC-BY-4.0 license.

arXiv:2011.12468

Nudge: Accelerating Overdue Pull Requests Towards Completion

Pull requests are a key part of the collaborative software development and code review process today. However, pull requests can also slow down the software development process when the reviewer(s) or the author do not actively engage with the pull request. In this work, we design an end-to-end service, Nudge, for accelerating overdue pull requests towards completion by reminding the author or the reviewer(s) to engage with their overdue pull requests. First, we use models based on effort estimation and machine learning to predict the completion time for a given pull request. Second, we use activity detection to filter out pull requests that may be overdue, but for which sufficient action is taking place nonetheless. Lastly, we use actor identification to understand who the blocker of the pull request is and nudge the appropriate actor (author or reviewer(s)). The key novelty of Nudge is that it succeeds in reducing pull request resolution time, while ensuring that developers perceive the notifications sent as useful, at the scale of thousands of repositories. In a randomized trial on 147 repositories in use at Microsoft, Nudge was able to reduce pull request resolution time by 60% for 8,500 pull requests, when compared to overdue pull requests for which Nudge did not send a notification. Furthermore, developers receiving Nudge notifications resolved 73% of these notifications as positive. We observed similar results when scaling up the deployment of Nudge to 8,000 repositories at Microsoft, for which Nudge sent 210,000 notifications during a full year. This demonstrates Nudge's ability to scale to thousands of repositories. Lastly, our qualitative analysis of a selection of Nudge notifications indicates areas for future research, such as taking dependencies among pull requests and developer availability into account.

arXiv:2011.08492

A Time-Frequency based Suspicious Activity Detection for Anti-Money Laundering

Money laundering is the crucial mechanism utilized by criminals to inject proceeds of crime to the financial system. The primary responsibility of the detection of suspicious activity related to money laundering is with the financial institutions. Most of the current systems in these institutions are rule-based and ineffective. The available data science-based anti-money laundering (AML) models in order to replace the existing rule-based systems work on customer relationship management (CRM) features and time characteristics of transaction behaviour. However, there is still a challenge on accuracy and problems around feature engineering due to thousands of possible features. Aiming to improve the detection performance of suspicious transaction monitoring systems for AML systems, in this article, we introduce a novel feature set based on time-frequency analysis, that makes use of 2-D representations of financial transactions. Random forest is utilized as a machine learning method, and simulated annealing is adopted for hyperparameter tuning. The designed algorithm is tested on real banking data, proving the efficacy of the results in practically relevant environments. It is shown that the time-frequency characteristics of suspicious and non-suspicious entities differentiate significantly, which would substantially improve the precision of data science-based transaction monitoring systems looking at only time-series transaction and CRM features.

arXiv:2010.14982

Toyota Smarthome Untrimmed: Real-World Untrimmed Videos for Activity Detection

Designing activity detection systems that can be successfully deployed in daily-living environments requires datasets that pose the challenges typical of real-world scenarios. In this paper, we introduce a new untrimmed daily-living dataset that features several real-world challenges: Toyota Smarthome Untrimmed (TSU). TSU contains a wide variety of activities performed in a spontaneous manner. The dataset contains dense annotations including elementary, composite activities and activities involving interactions with objects. We provide an analysis of the real-world challenges featured by our dataset, highlighting the open issues for detection algorithms. We show that current state-of-the-art methods fail to achieve satisfactory performance on the TSU dataset. Therefore, we propose a new baseline method for activity detection to tackle the novel challenges provided by our dataset. This method leverages one modality (i.e. optic flow) to generate the attention weights to guide another modality (i.e RGB) to better detect the activity boundaries. This is particularly beneficial to detect activities characterized by high temporal variance. We show that the method we propose outperforms state-of-the-art methods on TSU and on another popular challenging dataset, Charades.

arXiv:2010.13288

Activity Detection And Modeling Using Smart Meter Data: Concept And Case Studies

Electricity consumed by residential consumers counts for a significant part of global electricity consumption and utility companies can collect high-resolution load data thanks to the widely deployed advanced metering infrastructure. There has been a growing research interest toward appliance load disaggregation via nonintrusive load monitoring. As the electricity consumption of appliances is directly associated with the activities of consumers, this paper proposes a new and more effective approach, i.e., activity disaggregation. We present the concept of activity disaggregation and discuss its advantage over traditional appliance load disaggregation. We develop a framework by leverage machine learning for activity detection based on residential load data and features. We show through numerical case studies to demonstrate the effectiveness of the activity detection method and analyze consumer behaviors by time-dependent activity modeling. Last but not least, we discuss some potential use cases that can benefit from activity disaggregation and some future research directions.

arXiv:2010.12025

Combination of Deep Speaker Embeddings for Diarisation

Significant progress has recently been made in speaker diarisation after the introduction of d-vectors as speaker embeddings extracted from neural network (NN) speaker classifiers for clustering speech segments. To extract better-performing and more robust speaker embeddings, this paper proposes a c-vector method by combining multiple sets of complementary d-vectors derived from systems with different NN components. Three structures are used to implement the c-vectors, namely 2D self-attentive, gated additive, and bilinear pooling structures, relying on attention mechanisms, a gating mechanism, and a low-rank bilinear pooling mechanism respectively. Furthermore, a neural-based single-pass speaker diarisation pipeline is also proposed in this paper, which uses NNs to achieve voice activity detection, speaker change point detection, and speaker embedding extraction. Experiments and detailed analyses are conducted on the challenging AMI and NIST RT05 datasets which consist of real meetings with 4--10 speakers and a wide range of acoustic conditions. For systems trained on the AMI training set, relative speaker error rate (SER) reductions of 13% and 29% are obtained by using c-vectors instead of d-vectors on the AMI dev and eval sets respectively, and a relative reduction of 15% in SER is observed on RT05, which shows the robustness of the proposed methods. By incorporating VoxCeleb data into the training set, the best c-vector system achieved 7%, 17% and16% relative SER reduction compared to the d-vector on the AMI dev, eval, and RT05 sets respectively

arXiv:2010.12484

Speech enhancement aided end-to-end multi-task learning for voice activity detection

Robust voice activity detection (VAD) is a challenging task in low signal-to-noise (SNR) environments. Recent studies show that speech enhancement is helpful to VAD, but the performance improvement is limited. To address this issue, here we propose a speech enhancement aided end-to-end multi-task model for VAD. The model has two decoders, one for speech enhancement and the other for VAD. The two decoders share the same encoder and speech separation network. Unlike the direct thought that takes two separated objectives for VAD and speech enhancement respectively, here we propose a new joint optimization objective -- VAD-masked scale-invariant source-to-distortion ratio (mSI-SDR). mSI-SDR uses VAD information to mask the output of the speech enhancement decoder in the training process. It makes the VAD and speech enhancement tasks jointly optimized not only at the shared encoder and separation network, but also at the objective level. It also satisfies real-time working requirement theoretically. Experimental results show that the multi-task method significantly outperforms its single-task VAD counterpart. Moreover, mSI-SDR outperforms SI-SDR in the same multi-task setting.

arXiv:2010.11657

The HUAWEI Speaker Diarisation System for the VoxCeleb Speaker Diarisation Challenge

This paper describes system setup of our submission to speaker diarisation track (Track 4) of VoxCeleb Speaker Recognition Challenge 2020. Our diarisation system consists of a well-trained neural network based speech enhancement model as pre-processing front-end of input speech signals. We replace conventional energy-based voice activity detection (VAD) with a neural network based VAD. The neural network based VAD provides more accurate annotation of speech segments containing only background music, noise, and other interference, which is crucial to diarisation performance. We apply agglomerative hierarchical clustering (AHC) of x-vectors and variational Bayesian hidden Markov model (VB-HMM) based iterative clustering for speaker clustering. Experimental results demonstrate that our proposed system achieves substantial improvements over the baseline system, yielding diarisation error rate (DER) of 10.45%, and Jacard error rate (JER) of 22.46% on the evaluation set.

arXiv:2010.05497

The "Sound of Silence" in EEG -- Cognitive voice activity detection

Speech cognition bears potential application as a brain computer interface that can improve the quality of life for the otherwise communication impaired people. While speech and resting state EEG are popularly studied, here we attempt to explore a "non-speech"(NS) state of brain activity corresponding to the silence regions of speech audio. Firstly, speech perception is studied to inspect the existence of such a state, followed by its identification in speech imagination. Analogous to how voice activity detection is employed to enhance the performance of speech recognition, the EEG state activity detection protocol implemented here is applied to boost the confidence of imagined speech EEG decoding. Classification of speech and NS state is done using two datasets collected from laboratory-based and commercial-based devices. The state sequential information thus obtained is further utilized to reduce the search space of imagined EEG unit recognition. Temporal signal structures and topographic maps of NS states are visualized across subjects and sessions. The recognition performance and the visual distinction observed demonstrates the existence of silence signatures in EEG.

arXiv:2010.02477

A Unified Deep Learning Framework for Short-Duration Speaker Verification in Adverse Environments

Speaker verification (SV) has recently attracted considerable research interest due to the growing popularity of virtual assistants. At the same time, there is an increasing requirement for an SV system: it should be robust to short speech segments, especially in noisy and reverberant environments. In this paper, we consider one more important requirement for practical applications: the system should be robust to an audio stream containing long non-speech segments, where a voice activity detection (VAD) is not applied. To meet these two requirements, we introduce feature pyramid module (FPM)-based multi-scale aggregation (MSA) and self-adaptive soft VAD (SAS-VAD). We present the FPM-based MSA to deal with short speech segments in noisy and reverberant environments. Also, we use the SAS-VAD to increase the robustness to long non-speech segments. To further improve the robustness to acoustic distortions (i.e., noise and reverberation), we apply a masking-based speech enhancement (SE) method. We combine SV, VAD, and SE models in a unified deep learning framework and jointly train the entire network in an end-to-end manner. To the best of our knowledge, this is the first work combining these three models in a deep learning framework. We conduct experiments on Korean indoor (KID) and VoxCeleb datasets, which are corrupted by noise and reverberation. The results show that the proposed method is effective for SV in the challenging conditions and performs better than the baseline i-vector and deep speaker embedding systems.

arXiv:2003.12266

Dual Attention in Time and Frequency Domain for Voice Activity Detection

Voice activity detection (VAD) is a challenging task in low signal-to-noise ratio (SNR) environment, especially in non-stationary noise. To deal with this issue, we propose a novel attention module that can be integrated in Long Short-Term Memory (LSTM). Our proposed attention module refines each LSTM layer's hidden states so as to make it possible to adaptively focus on both time and frequency domain. Experiments are conducted on various noisy conditions using Aurora 4 database. Our proposed method obtains the 95.58 % area under the ROC curve (AUC), achieving 22.05 % relative improvement compared to baseline, with only 2.44 % increase in the number of parameters. Besides, we utilize focal loss for alleviating the performance degradation caused by imbalance between speech and non-speech sections in training sets. The results show that the focal loss can improve the performance in various imbalance situations compared to the cross entropy loss, a commonly used loss function in VAD.

arXiv:2009.11204

Learning Visual Voice Activity Detection with an Automatically Annotated Dataset

Visual voice activity detection (V-VAD) uses visual features to predict whether a person is speaking or not. V-VAD is useful whenever audio VAD (A-VAD) is inefficient either because the acoustic signal is difficult to analyze or because it is simply missing. We propose two deep architectures for V-VAD, one based on facial landmarks and one based on optical flow. Moreover, available datasets, used for learning and for testing V-VAD, lack content variability. We introduce a novel methodology to automatically create and annotate very large datasets in-the-wild -- WildVVAD -- based on combining A-VAD with face detection and tracking. A thorough empirical evaluation shows the advantage of training the proposed deep V-VAD models with this dataset.

arXiv:2009.09984

TRECVID 2019: An Evaluation Campaign to Benchmark Video Activity Detection, Video Captioning and Matching, and Video Search & Retrieval

The TREC Video Retrieval Evaluation (TRECVID) 2019 was a TREC-style video analysis and retrieval evaluation, the goal of which remains to promote progress in research and development of content-based exploitation and retrieval of information from digital video via open, metrics-based evaluation. Over the last nineteen years this effort has yielded a better understanding of how systems can effectively accomplish such processing and how one can reliably benchmark their performance. TRECVID has been funded by NIST (National Institute of Standards and Technology) and other US government agencies. In addition, many organizations and individuals worldwide contribute significant time and effort. TRECVID 2019 represented a continuation of four tasks from TRECVID 2018. In total, 27 teams from various research organizations worldwide completed one or more of the following four tasks: 1. Ad-hoc Video Search (AVS) 2. Instance Search (INS) 3. Activities in Extended Video (ActEV) 4. Video to Text Description (VTT) This paper is an introduction to the evaluation framework, tasks, data, and measures used in the workshop.

arXiv:2009.05527

On Multitask Loss Function for Audio Event Detection and Localization

Audio event localization and detection (SELD) have been commonly tackled using multitask models. Such a model usually consists of a multi-label event classification branch with sigmoid cross-entropy loss for event activity detection and a regression branch with mean squared error loss for direction-of-arrival estimation. In this work, we propose a multitask regression model, in which both (multi-label) event detection and localization are formulated as regression problems and use the mean squared error loss homogeneously for model training. We show that the common combination of heterogeneous loss functions causes the network to underfit the data whereas the homogeneous mean squared error loss leads to better convergence and performance. Experiments on the development and validation sets of the DCASE 2020 SELD task demonstrate that the proposed system also outperforms the DCASE 2020 SELD baseline across all the detection and localization metrics, reducing the overall SELD error (the combined metric) by approximately 10% absolute.

arXiv:2008.09466공식 코드 · Papers with Code

RespVAD: Voice Activity Detection via Video-Extracted Respiration Patterns

Voice Activity Detection (VAD) refers to the task of identification of regions of human speech in digital signals such as audio and video. While VAD is a necessary first step in many speech processing systems, it poses challenges when there are high levels of ambient noise during the audio recording. To improve the performance of VAD in such conditions, several methods utilizing the visual information extracted from the region surrounding the mouth/lip region of the speakers' video recording have been proposed. Even though these provide advantages over audio-only methods, they depend on faithful extraction of lip/mouth regions. Motivated by these, a new paradigm for VAD based on the fact that respiration forms the primary source of energy for speech production is proposed. Specifically, an audio-independent VAD technique using the respiration pattern extracted from the speakers' video is developed. The Respiration Pattern is first extracted from the video focusing on the abdominal-thoracic region of a speaker using an optical flow based method. Subsequently, voice activity is detected from the respiration pattern signal using neural sequence-to-sequence prediction models. The efficacy of the proposed method is demonstrated through experiments on a challenging dataset recorded in real acoustic environments and compared with four previous methods based on audio and visual cues.

arXiv:2008.08452

SegCodeNet: Color-Coded Segmentation Masks for Activity Detection from Wearable Cameras

Activity detection from first-person videos (FPV) captured using a wearable camera is an active research field with potential applications in many sectors, including healthcare, law enforcement, and rehabilitation. State-of-the-art methods use optical flow-based hybrid techniques that rely on features derived from the motion of objects from consecutive frames. In this work, we developed a two-stream network, the \emph{SegCodeNet}, that uses a network branch containing video-streams with color-coded semantic segmentation masks of relevant objects in addition to the original RGB video-stream. We also include a stream-wise attention gating that prioritizes between the two streams and a frame-wise attention module that prioritizes the video frames that contain relevant features. Experiments are conducted on an FPV dataset containing $18$ activity classes in office environments. In comparison to a single-stream network, the proposed two-stream method achieves an absolute improvement of $14.366\%$ and $10.324\%$ for averaged F1 score and accuracy, respectively, when average results are compared for three different frame sizes $224\times224$, $112\times112$, and $64\times64$. The proposed method provides significant performance gains for lower-resolution images with absolute improvements of $17\%$ and $26\%$ in F1 score for input dimensions of $112\times112$ and $64\times64$, respectively. The best performance is achieved for a frame size of $224\times224$ yielding an F1 score and accuracy of $90.176\%$ and $90.799\%$ which outperforms the state-of-the-art Inflated 3D ConvNet (I3D) \cite{carreira2017quo} method by an absolute margin of $4.529\%$ and $2.419\%$, respectively.