PyTorchKR Papers

최신 논문

총 361편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2103.10661

USTC-NELSLIP System Description for DIHARD-III Challenge

This system description describes our submission system to the Third DIHARD Speech Diarization Challenge. Besides the traditional clustering based system, the innovation of our system lies in the combination of various front-end techniques to solve the diarization problem, including speech separation and target-speaker based voice activity detection (TS-VAD), combined with iterative data purification. We also adopted audio domain classification to design domain-dependent processing. Finally, we performed post processing to do system fusion and selection. Our best system achieved DERs of 11.30% in track 1 and 16.78% in track 2 on evaluation set, respectively.

arXiv:2103.07125공식 코드 · Papers with Code

Learning spectro-temporal representations of complex sounds with parameterized neural networks

Deep Learning models have become potential candidates for auditory neuroscience research, thanks to their recent successes on a variety of auditory tasks. Yet, these models often lack interpretability to fully understand the exact computations that have been performed. Here, we proposed a parametrized neural network layer, that computes specific spectro-temporal modulations based on Gabor kernels (Learnable STRFs) and that is fully interpretable. We evaluated predictive capabilities of this layer on Speech Activity Detection, Speaker Verification, Urban Sound Classification and Zebra Finch Call Type Classification. We found out that models based on Learnable STRFs are on par for all tasks with different toplines, and obtain the best performance for Speech Activity Detection. As this layer is fully interpretable, we used quantitative measures to describe the distribution of the learned spectro-temporal modulations. The filters adapted to each task and focused mostly on low temporal and spectral modulations. The analyses show that the filters learned on human speech have similar spectro-temporal parameters as the ones measured directly in the human auditory cortex. Finally, we observed that the tasks organized in a meaningful way: the human vocalizations tasks closer to each other and bird vocalizations far away from human vocalizations and urban sounds tasks.

arXiv:2103.04792

An Ultra-low Power RNN Classifier for Always-On Voice Wake-Up Detection Robust to Real-World Scenarios

We present in this paper an ultra-low power (ULP) Recurrent Neural Network (RNN) based classifier for an always-on voice Wake-Up Sensor (WUS) with performances suitable for real-world applications. The purpose of our sensor is to bring down by at least a factor 100 the power consumption in background noise of always-on speech processing algorithms such as Automatic Speech Recognition, Keyword Spotting, Speaker Verification, etc. Unlike the other published approaches, we designed our wake-up sensor to be robust to unseen real-world noises for realistic levels of speech and noise by carefully designing the dataset and the loss function. We also specifically trained it to mark only the speech start rather than adopting a traditional Voice Activity Detection (VAD) approach. We achieve less than 3% No Trigger Rate (NTR) for a duty cycle less than 1% in challenging background noises pooled using a model of an analogue front-end. We demonstrate the superiority of RNNs on this task compared to the other tested approaches, with an estimated power consumption of 45 nW for the RNN itself in 65nm CMOS and a minimal memory footprint of 0.52 kB.

arXiv:2103.01302공식 코드 · Papers with Code

Coarse-Fine Networks for Temporal Activity Detection in Videos

In this paper, we introduce Coarse-Fine Networks, a two-stream architecture which benefits from different abstractions of temporal resolution to learn better video representations for long-term motion. Traditional Video models process inputs at one (or few) fixed temporal resolution without any dynamic frame selection. However, we argue that, processing multiple temporal resolutions of the input and doing so dynamically by learning to estimate the importance of each frame can largely improve video representations, specially in the domain of temporal activity localization. To this end, we propose (1) Grid Pool, a learned temporal downsampling layer to extract coarse features, and, (2) Multi-stage Fusion, a spatio-temporal attention mechanism to fuse a fine-grained context with the coarse features. We show that our method outperforms the state-of-the-arts for action detection in public datasets including Charades with a significantly reduced compute and memory footprint. The code is available at https://github.com/kkahatapitiya/Coarse-Fine-Networks

arXiv:2103.01661

Incorporating VAD into ASR System by Multi-task Learning

When we use End-to-end automatic speech recognition (E2E-ASR) system for real-world applications, a voice activity detection (VAD) system is usually needed to improve the performance and to reduce the computational cost by discarding non-speech parts in the audio. Usually ASR and VAD systems are trained and utilized independently to each other. In this paper, we present a novel multi-task learning (MTL) framework that incorporates VAD into the ASR system. The proposed system learns ASR and VAD jointly in the training stage. With the assistance of VAD, the ASR performance improves as its connectionist temporal classification (CTC) loss function can leverage the VAD alignment information. In the inference stage, the proposed system removes non-speech parts at low computational cost and recognizes speech parts with high robustness. Experimental results on segmented speech data show that by utilizing VAD information, the proposed method outperforms the baseline ASR system on both English and Chinese datasets. On unsegmented speech data, we find that the system outperforms the ASR systems that build an extra GMM-based or DNN-based voice activity detector.

arXiv:2102.11585공식 코드 · Papers with Code

ROAD: The ROad event Awareness Dataset for Autonomous Driving

Humans drive in a holistic fashion which entails, in particular, understanding dynamic road events and their evolution. Injecting these capabilities in autonomous vehicles can thus take situational awareness and decision making closer to human-level performance. To this purpose, we introduce the ROad event Awareness Dataset (ROAD) for Autonomous Driving, to our knowledge the first of its kind. ROAD is designed to test an autonomous vehicle's ability to detect road events, defined as triplets composed by an active agent, the action(s) it performs and the corresponding scene locations. ROAD comprises videos originally from the Oxford RobotCar Dataset annotated with bounding boxes showing the location in the image plane of each road event. We benchmark various detection tasks, proposing as a baseline a new incremental algorithm for online road event awareness termed 3D-RetinaNet. We also report the performance on the ROAD tasks of Slowfast and YOLOv5 detectors, as well as that of the winners of the ICCV2021 ROAD challenge, which highlight the challenges faced by situation awareness in autonomous driving. ROAD is designed to allow scholars to investigate exciting tasks such as complex (road) activity detection, future event anticipation and continual learning. The dataset is available at https://github.com/gurkirt/road-dataset; the baseline can be found at https://github.com/gurkirt/3D-RetinaNet.

arXiv:2101.03478

Activity Recognition with Moving Cameras and Few Training Examples: Applications for Detection of Autism-Related Headbanging

Activity recognition computer vision algorithms can be used to detect the presence of autism-related behaviors, including what are termed "restricted and repetitive behaviors", or stimming, by diagnostic instruments. The limited data that exist in this domain are usually recorded with a handheld camera which can be shaky or even moving, posing a challenge for traditional feature representation approaches for activity detection which mistakenly capture the camera's motion as a feature. To address these issues, we first document the advantages and limitations of current feature representation techniques for activity recognition when applied to head banging detection. We then propose a feature representation consisting exclusively of head pose keypoints. We create a computer vision classifier for detecting head banging in home videos using a time-distributed convolutional neural network (CNN) in which a single CNN extracts features from each frame in the input sequence, and these extracted features are fed as input to a long short-term memory (LSTM) network. On the binary task of predicting head banging and no head banging within videos from the Self Stimulatory Behaviour Dataset (SSBD), we reach a mean F1-score of 90.77% using 3-fold cross validation (with individual fold F1-scores of 83.3%, 89.0%, and 100.0%) when ensuring that no child who appeared in the train set was in the test set for all folds. This work documents a successful technique for training a computer vision classifier which can detect human motion with few training examples and even when the camera recording the source clips is unstable. The general methods described here can be applied by designers and developers of interactive systems towards other human motion and pose classification problems used in mobile and ubiquitous interactive systems.

arXiv:2011.12998공식 코드 · Papers with Code

VOXLINGUA107: A DATASET FOR SPOKEN LANGUAGE RECOGNITION

This paper investigates the use of automatically collected web audio data for the task of spoken language recognition. We generate semi-random search phrases from language-specific Wikipedia data that are then used to retrieve videos from YouTube for 107 languages. Speech activity detection and speaker diarization are used to extract segments from the videos that contain speech. Post-filtering is used to remove segments from the database that are likely not in the given language, increasing the proportion of correctly labeled segments to 98%, based on crowd-sourced verification. The size of the resulting training set (VoxLingua107) is 6628 hours (62 hours per language on the average) and it is accompanied by an evaluation set of 1609 verified utterances. We use the data to build language recognition models for several spoken language identification tasks. Experiments show that using the automatically retrieved training data gives competitive results to using hand-labeled proprietary datasets. The dataset is publicly available.

arXiv:2012.07467공식 코드 · Papers with Code

AV Taris: Online Audio-Visual Speech Recognition

In recent years, Automatic Speech Recognition (ASR) technology has approached human-level performance on conversational speech under relatively clean listening conditions. In more demanding situations involving distant microphones, overlapped speech, background noise, or natural dialogue structures, the ASR error rate is at least an order of magnitude higher. The visual modality of speech carries the potential to partially overcome these challenges and contribute to the sub-tasks of speaker diarisation, voice activity detection, and the recovery of the place of articulation, and can compensate for up to 15dB of noise on average. This article develops AV Taris, a fully differentiable neural network model capable of decoding audio-visual speech in real time. We achieve this by connecting two recently proposed models for audio-visual speech integration and online speech recognition, namely AV Align and Taris. We evaluate AV Taris under the same conditions as AV Align and Taris on one of the largest publicly available audio-visual speech datasets, LRS2. Our results show that AV Taris is superior to the audio-only variant of Taris, demonstrating the utility of the visual modality to speech recognition within the real time decoding framework defined by Taris. Compared to an equivalent Transformer-based AV Align model that takes advantage of full sentences without meeting the real-time requirement, we report an absolute degradation of approximately 3% with AV Taris. As opposed to the more popular alternative for online speech recognition, namely the RNN Transducer, Taris offers a greatly simplified fully differentiable training pipeline. As a consequence, AV Taris has the potential to popularise the adoption of Audio-Visual Speech Recognition (AVSR) technology and overcome the inherent limitations of the audio modality in less optimal listening conditions.

arXiv:2009.02089

Massive Machine Type Communication Pilot-Hopping Sequence Detection Architectures Based on Non-Negative Least Squares for Grant-Free Random Access

User activity detection in grant-free random access massive machine type communication (mMTC) using pilot-hopping sequences can be formulated as solving a non-negative least squares (NNLS) problem. In this work, two architectures using different algorithms to solve the NNLS problem is proposed. The algorithms are implemented using a fully parallel approach and fixed-point arithmetic, leading to high detection rates and low power consumption. The first algorithm, fast projected gradients, converges faster to the optimal value. The second algorithm, multiplicative updates, is partially implemented in the logarithmic domain, and provides a smaller chip area and lower power consumption. For a detection rate of about one million detections per second, the chip area for the fast algorithm is about 0.7 mm$^2$ compared to about 0.5 mm$^2$ for the multiplicative algorithm when implemented in a 28 nm FD-SOI standard cell process at 1 V power supply voltage. The energy consumption is about 300 nJ/detection for the fast projected gradient algorithm using 256 iterations, leading to a convergence close to the theoretical. With 128 iterations, about 250 nJ/detection is required, with a detection performance on par with 192 iterations of the multiplicative algorithm for which about 100 nJ/detection is required.

arXiv:2010.13886

MarbleNet: Deep 1D Time-Channel Separable Convolutional Neural Network for Voice Activity Detection

We present MarbleNet, an end-to-end neural network for Voice Activity Detection (VAD). MarbleNet is a deep residual network composed from blocks of 1D time-channel separable convolution, batch-normalization, ReLU and dropout layers. When compared to a state-of-the-art VAD model, MarbleNet is able to achieve similar performance with roughly 1/10-th the parameter cost. We further conduct extensive ablation studies on different training methods and choices of parameters in order to study the robustness of MarbleNet in real-world VAD tasks.

arXiv:2010.12692

Multi-Channel Speaker Verification for Single and Multi-talker Speech

To improve speaker verification in real scenarios with interference speakers, noise, and reverberation, we propose to bring together advancements made in multi-channel speech features. Specifically, we combine spectral, spatial, and directional features, which includes inter-channel phase difference, multi-channel sinc convolutions, directional power ratio features, and angle features. To maximally leverage supervised learning, our framework is also equipped with multi-channel speech enhancement and voice activity detection. On all simulated, replayed, and real recordings, we observe large and consistent improvements at various degradation levels. On real recordings of multi-talker speech, we achieve a 36% relative reduction in equal error rate w.r.t. single-channel baseline. We find the improvements from speaker-dependent directional features more consistent in multi-talker conditions than clean. Lastly, we investigate if the learned multi-channel speaker embedding space can be made more discriminative through a contrastive loss-based fine-tuning. With a simple choice of Triplet loss, we observe a further 8.3% relative reduction in EER.

arXiv:2010.09979

An Efficient Algorithm for Device Detection and Channel Estimation in Asynchronous IoT Systems

A great amount of endeavour has recently been devoted to the joint device activity detection and channel estimation problem in massive machine-type communications. This paper targets at two practical issues along this line that have not been addressed before: asynchronous transmission from uncoordinated users and efficient algorithms for real-time implementation in systems with a massive number of devices. Specifically, this paper considers a practical system where the preamble sent by each active device is delayed by some unknown number of symbols due to the lack of coordination. We manage to cast the problem of detecting the active devices and estimating their delay and channels into a group LASSO problem. Then, a block coordinate descent algorithm is proposed to solve this problem globally, where the closed-form solution is available when updating each block of variables with the other blocks of variables being fixed, thanks to the special structure of our interested problem. Our analysis shows that the overall complexity of the proposed algorithm is low, making it suitable for real-time application.

arXiv:2009.12863

Grant-Free Access via Bilinear Inference for Cell-Free MIMO with Low-Coherent Pilots

We propose a novel joint activity, channel, and data estimation (JACDE) scheme for cell-free multiple-input multiple-output (MIMO) systems compliant with fifth-generation (5G) new radio (NR) orthogonal frequency-division multiplexing (OFDM) signaling. The contribution aims to allow significant overhead reduction of cell-free MIMO systems by enabling grant-free access, while maintaining moderate throughput per user. To that end, we extend the conventional MIMO OFDM protocol so as to incorporate activity detection capability without resorting to spreading informative data symbols, in contrast with related work which typically relies on signal spreading. Our method leverages a Bayesian message passing scheme based on Gaussian approximation, which jointly performs active user detection (AUD), channel estimation (CE), and multi-user detection (MUD), incorporating also a well-structured low-coherent pilot design based on frame theory, which mitigates pilot contamination, and finally complemented with a detector empowered by bilinear message passing. The efficacy of the resulting JACDE-based grant-free access scheme without spreading data sequences is demonstrated by simulation results, which are shown to significantly outperform the current state-of-the-art and approach the performance of an idealized (genie-aided) scheme in which user activity and channel coefficients are perfectly known.

arXiv:2006.07898

The JHU Multi-Microphone Multi-Speaker ASR System for the CHiME-6 Challenge

This paper summarizes the JHU team's efforts in tracks 1 and 2 of the CHiME-6 challenge for distant multi-microphone conversational speech diarization and recognition in everyday home environments. We explore multi-array processing techniques at each stage of the pipeline, such as multi-array guided source separation (GSS) for enhancement and acoustic model training data, posterior fusion for speech activity detection, PLDA score fusion for diarization, and lattice combination for automatic speech recognition (ASR). We also report results with different acoustic model architectures, and integrate other techniques such as online multi-channel weighted prediction error (WPE) dereverberation and variational Bayes-hidden Markov model (VB-HMM) based overlap assignment to deal with reverberation and overlapping speakers, respectively. As a result of these efforts, our ASR systems achieve a word error rate of 40.5% and 67.5% on tracks 1 and 2, respectively, on the evaluation set. This is an improvement of 10.8% and 10.4% absolute, over the challenge baselines for the respective tracks.

arXiv:2006.05637

Distributed Optimization for Massive Connectivity

Massive device connectivity in Internet of Thing (IoT) networks with sporadic traffic poses significant communication challenges. To overcome this challenge, the serving base station is required to detect the active devices and estimate the corresponding channel state information during each coherence block. The corresponding joint activity detection and channel estimation problem can be formulated as a group sparse estimation problem, also known under the name "Group Lasso". This letter presents a fast and efficient distributed algorithm to solve such Group Lasso problems, which alternates between solving small-scaled problems in parallel and dealing with a linear equation for consensus. Numerical results demonstrate the speedup of this algorithm compared with the state-of-the-art methods in terms of convergence speed and computation time.

arXiv:2005.10145

Real-Time Radar-Based Gesture Detection and Recognition Built in an Edge-Computing Platform

In this paper, a real-time signal processing frame-work based on a 60 GHz frequency-modulated continuous wave (FMCW) radar system to recognize gestures is proposed. In order to improve the robustness of the radar-based gesture recognition system, the proposed framework extracts a comprehensive hand profile, including range, Doppler, azimuth and elevation, over multiple measurement-cycles and encodes them into a feature cube. Rather than feeding the range-Doppler spectrum sequence into a deep convolutional neural network (CNN) connected with recurrent neural networks, the proposed framework takes the aforementioned feature cube as input of a shallow CNN for gesture recognition to reduce the computational complexity. In addition, we develop a hand activity detection (HAD) algorithm to automatize the detection of gestures in real-time case. The proposed HAD can capture the time-stamp at which a gesture finishes and feeds the hand profile of all the relevant measurement-cycles before this time-stamp into the CNN with low latency. Since the proposed framework is able to detect and classify gestures at limited computational cost, it could be deployed in an edge-computing platform for real-time applications, whose performance is notedly inferior to a state-of-the-art personal computer. The experimental results show that the proposed framework has the capability of classifying 12 gestures in real-time with a high F1-score.

arXiv:2005.09913

Statistical and Neural Network Based Speech Activity Detection in Non-Stationary Acoustic Environments

Speech activity detection (SAD), which often rests on the fact that the noise is "more" stationary than speech, is particularly challenging in non-stationary environments, because the time variance of the acoustic scene makes it difficult to discriminate speech from noise. We propose two approaches to SAD, where one is based on statistical signal processing, while the other utilizes neural networks. The former employes sophisticated signal processing to track the noise and speech energies and is meant to support the case for a resource efficient, unsupervised signal processing approach. The latter introduces a recurrent network layer that operates on short segments of the input speech to do temporal smoothing in the presence of non-stationary noise. The systems are tested on the Fearless Steps challenge, which consists of the transmission data from the Apollo-11 space mission. The statistical SAD achieves comparable detection performance to earlier proposed neural network based SADs, while the neural network based approach leads to a decision cost function of 1.07% on the evaluation set of the 2020 Fearless Steps Challenge, which sets a new state of the art.

arXiv:1912.10442

End-Point Detection with State Transition Model based on Chunk-Wise Classification

A state transition model (STM) based on chunk-wise classification was proposed for end-point detection (EPD). In general, EPD is developed using frame-wise voice activity detection (VAD) with additional STM, in which the state transition is conducted based on VAD's frame-level decision (speech or non-speech). However, VAD errors frequently occur in noisy environments, even though we use state-of-the-art deep neural network based VAD, which causes the undesired state transition of STM. In this work, to build robust STM, a state transition is conducted based on chunk-wise classification as EPD does not need to be conducted in frame-level. The chunk consists of multiple frames and the classification of chunk between speech and non-speech is done by aggregating the decisions of VAD for multiple frames, so that some undesired VAD errors in a chunk can be smoothed by other correct VAD decisions. Finally, the model was evaluated in both qualitative and quantitative measures including phone error rate.

arXiv:1911.12157

Intelligent Reflecting Surface for Massive Device Connectivity: Joint Activity Detection and Channel Estimation

Intelligent Reflecting Surface (IRS) has been a promising solution to enhance wireless networks both spectral-efficiently and energy-efficiently. This paper considers an IRS-assisted the Internet of Things network for massive connectivity. We aim to solve the IRS-related activity detection and channel estimation problem which has not been studied before. In this paper, we formulate the IRS-related activity detection and channel estimation problem as sparse matrix factorization, matrix completion and Multiple Measurement Vector problem and, we propose a three-stage framework based on the approximate message passing. Simulation results verify the effectiveness of the proposed algorithm.