arXiv subjects

Harishchandra Dubey

Publications and source records attributed to Harishchandra Dubey.

At least 19 recordsLinked to original sources

arXiv2023

ICASSP 2023 Deep Noise Suppression Challenge

Deep Speech Enhancement Challenge is the 5th edition of deep noise suppression (DNS) challenges organized at ICASSP 2023 Signal Processing Grand Challenges. DNS challenges were organized during 2019-2023 to stimulate research in deep speech enhancement (DSE). Previous DNS challenges were organized at INTERSPEECH 2020, ICASSP 2021, INTERSPEECH 2021, and ICASSP 2022. From prior editions, we learnt that improving signal quality (SIG) is challenging particularly in presence of simultaneously active interfering talkers and noise. This challenge aims to develop models for joint denosing, dereverberation and suppression of interfering talkers. When primary talker wears a headphone, certain acoustic properties of their speech such as direct-to-reverberation (DRR), signal to noise ratio (SNR) etc. make it possible to suppress neighboring talkers even without enrollment data for primary talker. This motivated us to create two tracks for this challenge: (i) Track-1 Headset; (ii) Track-2 Speakerphone. Both tracks has fullband (48kHz) training data and testset, and each testclips has a corresponding enrollment data (10-30s duration) for primary talker. Each track invited submissions of personalized and non-personalized models all of which are evaluated through same subjective evaluation. Most models submitted to challenge were personalized models, same team is winner in both tracks where the best models has improvement of 0.145 and 0.141 in challenge's Score as compared to noisy blind testset.

cs.SD↗

arXiv2022

MusicNet: Compact Convolutional Neural Network for Real-time Background Music Detection

With the recent growth of remote work, online meetings often encounter challenging audio contexts such as background noise, music, and echo. Accurate real-time detection of music events can help to improve the user experience. In this paper, we present MusicNet, a compact neural model for detecting background music in the real-time communications pipeline. In video meetings, music frequently co-occurs with speech and background noises, making the accurate classification quite challenging. We propose a compact convolutional neural network core preceded by an in-model featurization layer. MusicNet takes 9 seconds of raw audio as input and does not require any model-specific featurization in the product stack. We train our model on the balanced subset of the Audio Set~\cite{gemmeke2017audio} data and validate it on 1000 crowd-sourced real test clips. Finally, we compare MusicNet performance with 20 state-of-the-art models. MusicNet has a true positive rate (TPR) of 81.3% at a 0.1% false positive rate (FPR), which is significantly better than state-of-the-art models included in our study. MusicNet is also 10x smaller and has 4x faster inference than the best performing models we benchmarked.

eess.AS↗

arXiv2022

ICASSP 2022 Deep Noise Suppression Challenge

The Deep Noise Suppression (DNS) challenge is designed to foster innovation in the area of noise suppression to achieve superior perceptual speech quality. This is the 4th DNS challenge, with the previous editions held at INTERSPEECH 2020, ICASSP 2021, and INTERSPEECH 2021. We open-source datasets and test sets for researchers to train their deep noise suppression models, as well as a subjective evaluation framework based on ITU-T P.835 to rate and rank-order the challenge entries. We provide access to DNSMOS P.835 and word accuracy (WAcc) APIs to challenge participants to help with iterative model improvements. In this challenge, we introduced the following changes: (i) Included mobile device scenarios in the blind test set; (ii) Included a personalized noise suppression track with baseline; (iii) Added WAcc as an objective metric; (iv) Included DNSMOS P.835; (v) Made the training datasets and test sets fullband (48 kHz). We use an average of WAcc and subjective scores P.835 SIG, BAK, and OVRL to get the final score for ranking the DNS models. We believe that as a research community, we still have a long way to go in achieving excellent speech quality in challenging noisy real-world scenarios.

eess.AS↗

arXiv2021

Interspeech 2021 Deep Noise Suppression Challenge

The Deep Noise Suppression (DNS) challenge is designed to foster innovation in the area of noise suppression to achieve superior perceptual speech quality. We recently organized a DNS challenge special session at INTERSPEECH and ICASSP 2020. We open-sourced training and test datasets for the wideband scenario. We also open-sourced a subjective evaluation framework based on ITU-T standard P.808, which was also used to evaluate participants of the challenge. Many researchers from academia and industry made significant contributions to push the field forward, yet even the best noise suppressor was far from achieving superior speech quality in challenging scenarios. In this version of the challenge organized at INTERSPEECH 2021, we are expanding both our training and test datasets to accommodate full band scenarios. The two tracks in this challenge will focus on real-time denoising for (i) wide band, and(ii) full band scenarios. We are also making available a reliable non-intrusive objective speech quality metric called DNSMOS for the participants to use during their development phase.

cs.SD↗

arXiv2020

ICASSP 2021 Deep Noise Suppression Challenge

The Deep Noise Suppression (DNS) challenge is designed to foster innovation in the area of noise suppression to achieve superior perceptual speech quality. We recently organized a DNS challenge special session at INTERSPEECH 2020. We open sourced training and test datasets for researchers to train their noise suppression models. We also open sourced a subjective evaluation framework and used the tool to evaluate and pick the final winners. Many researchers from academia and industry made significant contributions to push the field forward. We also learned that as a research community, we still have a long way to go in achieving excellent speech quality in challenging noisy real-time conditions. In this challenge, we are expanding both our training and test datasets. There are two tracks with one focusing on real-time denoising and the other focusing on real-time personalized deep noise suppression. We also make a non-intrusive objective speech quality metric called DNSMOS available for participants to use during their development stages. The final evaluation will be based on subjective tests.

eess.AS↗

arXiv2020

The INTERSPEECH 2020 Deep Noise Suppression Challenge: Datasets, Subjective Testing Framework, and Challenge Results

The INTERSPEECH 2020 Deep Noise Suppression (DNS) Challenge is intended to promote collaborative research in real-time single-channel Speech Enhancement aimed to maximize the subjective (perceptual) quality of the enhanced speech. A typical approach to evaluate the noise suppression methods is to use objective metrics on the test set obtained by splitting the original dataset. While the performance is good on the synthetic test set, often the model performance degrades significantly on real recordings. Also, most of the conventional objective metrics do not correlate well with subjective tests and lab subjective tests are not scalable for a large test set. In this challenge, we open-sourced a large clean speech and noise corpus for training the noise suppression models and a representative test set to real-world scenarios consisting of both synthetic and real recordings. We also open-sourced an online subjective test framework based on ITU-T P.808 for researchers to reliably test their developments. We evaluated the results using P.808 on a blind test set. The results and the key learnings from the challenge are discussed. The datasets and scripts can be found here for quick access https://github.com/microsoft/DNS-Challenge.

eess.AS↗

arXiv2020

The INTERSPEECH 2020 Deep Noise Suppression Challenge: Datasets, Subjective Speech Quality and Testing Framework

The INTERSPEECH 2020 Deep Noise Suppression Challenge is intended to promote collaborative research in real-time single-channel Speech Enhancement aimed to maximize the subjective (perceptual) quality of the enhanced speech. A typical approach to evaluate the noise suppression methods is to use objective metrics on the test set obtained by splitting the original dataset. Many publications report reasonable performance on the synthetic test set drawn from the same distribution as that of the training set. However, often the model performance degrades significantly on real recordings. Also, most of the conventional objective metrics do not correlate well with subjective tests and lab subjective tests are not scalable for a large test set. In this challenge, we open-source a large clean speech and noise corpus for training the noise suppression models and a representative test set to real-world scenarios consisting of both synthetic and real recordings. We also open source an online subjective test framework based on ITU-T P.808 for researchers to quickly test their developments. The winners of this challenge will be selected based on subjective evaluation on a representative test set using P.808 framework.

cs.SD↗

arXiv2020

Weighted Speech Distortion Losses for Neural-network-based Real-time Speech Enhancement

This paper investigates several aspects of training a RNN (recurrent neural network) that impact the objective and subjective quality of enhanced speech for real-time single-channel speech enhancement. Specifically, we focus on a RNN that enhances short-time speech spectra on a single-frame-in, single-frame-out basis, a framework adopted by most classical signal processing methods. We propose two novel mean-squared-error-based learning objectives that enable separate control over the importance of speech distortion versus noise reduction. The proposed loss functions are evaluated by widely accepted objective quality and intelligibility measures and compared to other competitive online methods. In addition, we study the impact of feature normalization and varying batch sequence lengths on the objective quality of enhanced speech. Finally, we show subjective ratings for the proposed approach and a state-of-the-art real-time RNN-based method.

eess.AS↗

arXiv2020

CURE Dataset: Ladder Networks for Audio Event Classification

Audio event classification is an important task for several applications such as surveillance, audio, video and multimedia retrieval etc. There are approximately 3M people with hearing loss who can't perceive events happening around them. This paper establishes the CURE dataset which contains curated set of specific audio events most relevant for people with hearing loss. We propose a ladder network based audio event classifier that utilizes 5s sound recordings derived from the Freesound project. We adopted the state-of-the-art convolutional neural network (CNN) embeddings as audio features for this task. We also investigate extreme learning machine (ELM) for event classification. In this study, proposed classifiers are compared with support vector machine (SVM) baseline. We propose signal and feature normalization that aims to reduce the mismatch between different recordings scenarios. Firstly, CNN is trained on weakly labeled Audioset data. Next, the pre-trained model is adopted as feature extractor for proposed CURE corpus. We incorporate ESC-50 dataset as second evaluation set. Results and discussions validate the superiority of Ladder network over ELM and SVM classifier in terms of robustness and increased classification accuracy. While Ladder network is robust to data mismatches, simpler SVM and ELM classifiers are sensitive to such mismatches, where the proposed normalization techniques can play an important role. Experimental studies with ESC-50 and CURE corpora elucidate the differences in dataset complexity and robustness offered by proposed approaches.

cs.SD↗

arXiv2019

Fuzzy C-Means Clustering and Sonification of HRV Features

Linear and non-linear measures of heart rate variability (HRV) are widely investigated as non-invasive indicators of health. Stress has a profound impact on heart rate, and different meditation techniques have been found to modulate heartbeat rhythm. This paper aims to explore the process of identifying appropriate metrices from HRV analysis for sonification. Sonification is a type of auditory display involving the process of mapping data to acoustic parameters. This work explores the use of auditory display in aiding the analysis of HRV leveraged by unsupervised machine learning techniques. Unsupervised clustering helps select the appropriate features to improve the sonification interpretability. Vocal synthesis sonification techniques are employed to increase comprehension and learnability of the processed data displayed through sound. These analyses are early steps in building a real-time sound-based biofeedback training system.

cs.HC↗

arXiv2019

Toeplitz Inverse Covariance based Robust Speaker Clustering for Naturalistic Audio Streams

Speaker diarization determines who spoke and when? in an audio stream. In this study, we propose a model-based approach for robust speaker clustering using i-vectors. The ivectors extracted from different segments of same speaker are correlated. We model this correlation with a Markov Random Field (MRF) network. Leveraging the advancements in MRF modeling, we used Toeplitz Inverse Covariance (TIC) matrix to represent the MRF correlation network for each speaker. This approaches captures the sequential structure of i-vectors (or equivalent speaker turns) belonging to same speaker in an audio stream. A variant of standard Expectation Maximization (EM) algorithm is adopted for deriving closed-form solution using dynamic programming (DP) and the alternating direction method of multiplier (ADMM). Our diarization system has four steps: (1) ground-truth segmentation; (2) i-vector extraction; (3) post-processing (mean subtraction, principal component analysis, and length-normalization) ; and (4) proposed speaker clustering. We employ cosine K-means and movMF speaker clustering as baseline approaches. Our evaluation data is derived from: (i) CRSS-PLTL corpus, and (ii) two meetings subset of the AMI corpus. Relative reduction in diarization error rate (DER) for CRSS-PLTL corpus is 43.22% using the proposed advancements as compared to baseline. For AMI meetings IS1000a and IS1003b, relative DER reduction is 29.37% and 9.21%, respectively.

cs.SD↗

arXiv2018

Robust Speaker Clustering using Mixtures of von Mises-Fisher Distributions for Naturalistic Audio Streams

Speaker Diarization (i.e. determining who spoke and when?) for multi-speaker naturalistic interactions such as Peer-Led Team Learning (PLTL) sessions is a challenging task. In this study, we propose robust speaker clustering based on mixture of multivariate von Mises-Fisher distributions. Our diarization pipeline has two stages: (i) ground-truth segmentation; (ii) proposed speaker clustering. The ground-truth speech activity information is used for extracting i-Vectors from each speechsegment. We post-process the i-Vectors with principal component analysis for dimension reduction followed by lengthnormalization. Normalized i-Vectors are high-dimensional unit vectors possessing discriminative directional characteristics. We model the normalized i-Vectors with a mixture model consisting of multivariate von Mises-Fisher distributions. K-means clustering with cosine distance is chosen as baseline approach. The evaluation data is derived from: (i) CRSS-PLTL corpus; and (ii) three-meetings subset of AMI corpus. The CRSSPLTL data contain audio recordings of PLTL sessions which is student-led STEM education paradigm. Proposed approach is consistently better than baseline leading to upto 44.48% and 53.68% relative improvements for PLTL and AMI corpus, respectively. Index Terms: Speaker clustering, von Mises-Fisher distribution, Peer-led team learning, i-Vector, Naturalistic Audio.

cs.SD↗

arXiv2018

SmartEAR: Smartwatch-based Unsupervised Learning for Multi-modal Signal Analysis in Opportunistic Sensing Framework

Wrist-bands such as smartwatches have become an unobtrusive interface for collecting physiological and contextual data from users. Smartwatches are being used for smart healthcare, telecare, and wellness monitoring. In this paper, we used data collected from the AnEAR framework leveraging smartwatches to gather and store physiological data from patients in naturalistic settings. This data included temperature, galvanic skin response (GSR), acceleration, and heart rate (HR). In particular, we focused on HR and acceleration, as these two modalities are often correlated. Since the data was unlabeled we relied on unsupervised learning for multi-modal signal analysis. We propose using k-means clustering, GMM clustering, and Self-Organizing maps based on Neural Networks for group the multi-modal data into homogeneous clusters. This strategy helped in discovering latent structures in our data.

cs.CY↗

arXiv2018

Robust Feature Clustering for Unsupervised Speech Activity Detection

In certain applications such as zero-resource speech processing or very-low resource speech-language systems, it might not be feasible to collect speech activity detection (SAD) annotations. However, the state-of-the-art supervised SAD techniques based on neural networks or other machine learning methods require annotated training data matched to the target domain. This paper establish a clustering approach for fully unsupervised SAD useful for cases where SAD annotations are not available. The proposed approach leverages Hartigan dip test in a recursive strategy for segmenting the feature space into prominent modes. Statistical dip is invariant to distortions that lends robustness to the proposed method. We evaluate the method on NIST OpenSAD 2015 and NIST OpenSAT 2017 public safety communications data. The results showed the superiority of proposed approach over the two-component GMM baseline. Index Terms: Clustering, Hartigan dip test, NIST OpenSAD, NIST OpenSAT, speech activity detection, zero-resource speech processing, unsupervised learning.

cs.SD↗

arXiv2018

Fog Computing based SDI Framework for Mineral Resources Information Infrastructure Management in India

Spatial Data Infrastructure (SDI) is an important concept for sharing spatial data across the web. With cumulative techniques with spatial cloud computing and fog computing, SDI has the greater potential and has been emerged as a tool for processing, analysis and transmission of spatial data. The Fog computing is a paradigm where Fog devices help to increase throughput and reduce latency at the edge of the client with respect to cloud computing environment. This paper proposed and developed a fog computing based SDI framework for mining analytics from spatial big data for mineral resources management in India. We built a prototype using Raspberry Pi, an embedded microprocessor. We validated by taking suitable case study of mineral resources management in India by doing preliminary analysis including overlay analysis. Results showed that fog computing hold a great promise for analysis of spatial data. We used open source GIS i.e. QGIS and QIS plugin for reducing the transmission to cloud from the fog node.

cs.DC↗

arXiv2017

FogGrid: Leveraging Fog Computing for Enhanced Smart Grid Network

The present manuscript concentrates on the application of Fog computing to a Smart Grid Network that comprises of a Distribution Generation System known as a Microgrid. It addresses features and advantages of a smart grid. Two computational methods for on-demand processing based on shared information resources is discussed. Fog Computing acts as an additional layer of computational and/or communication nodes that offload the Cloud backend from multi-tasking while dealing with large amounts of data. Both Fog computing and Cloud computing hierarchical architecture is compared with respect to efficient utilization of resources. To alleviate the advantages of Fog computing, a Fog computing framework based on Intel Edison is proposed. The proposed architecture has been hardware implemented for a microgrid system. The results obtained show the efficacy of Fog Computing for smart grid network in terms of low power consumption, reduced storage requirement and overlay analysis capabilities.

cs.DC↗

arXiv2017

SoA-Fog: Secure Service-Oriented Edge Computing Architecture for Smart Health Big Data Analytics

The smart health paradigms employ Internet-connected wearables for telemonitoring, diagnosis for providing inexpensive healthcare solutions. Fog computing reduces latency and increases throughput by processing data near the body sensor network. In this paper, we proposed a secure serviceorientated edge computing architecture that is validated on recently released public dataset. Results and discussions support the applicability of proposed architecture for smart health applications. We proposed SoA-Fog i.e. a three-tier secure framework for efficient management of health data using fog devices. It discuss the security aspects in client layer, fog layer and the cloud layer. We design the prototype by using win-win spiral model with use case and sequence diagram. Overlay analysis was performed using proposed framework on malaria vector borne disease positive maps of Maharastra state in India from 2011 to 2014. The mobile clients were taken as test case. We performed comparative analysis between proposed secure fog framework and state-of-the art cloud-based framework.

cs.DC↗

arXiv2017

Smart Fog: Fog Computing Framework for Unsupervised Clustering Analytics in Wearable Internet of Things

The increasing use of wearables in smart telehealth generates heterogeneous medical big data. Cloud and fog services process these data for assisting clinical procedures. IoT based ehealthcare have greatly benefited from efficient data processing. This paper proposed and evaluated use of low resource machine learning on Fog devices kept close to the wearables for smart healthcare. In state of the art telecare systems, the signal processing and machine learning modules are deployed in the cloud for processing physiological data. We developed a prototype of Fog-based unsupervised machine learning big data analysis for discovering patterns in physiological data. We employed Intel Edison and Raspberry Pi as Fog computer in proposed architecture. We performed validation studies on real-world pathological speech data from in home monitoring of patients with Parkinson's disease (PD). Proposed architecture employed machine learning for analysis of pathological speech data obtained from smartwatches worn by the patients with PD. Results showed that proposed architecture is promising for low-resource clinical machine learning. It could be useful for other applications within wearable IoT for smart telehealth scenarios by translating machine learning approaches from the cloud backend to edge computing devices such as Fog.

cs.CY↗