Network


Latest external collaboration on country level. Dive into details by clicking on the dots.

Hotspot


Dive into the research topics where Hemant Misra is active.

Publication


Featured researches published by Hemant Misra.


international conference on acoustics, speech, and signal processing | 2003

New entropy based combination rules in HMM/ANN multi-stream ASR

Hemant Misra; Vivek Tyagi

Classifier performance is often enhanced through combining multiple streams of information. In the context of multi-stream HMM/ANN systems in ASR, a confidence measure widely used in classifier combination is the entropy of the posteriors distribution output from each ANN, which generally increases as classification becomes less reliable. The rule most commonly used is to select the ANN with the minimum entropy. However, this is not necessarily the best way to use entropy in classifier combination. In this article, we test three new entropy based combination rules in a full-combination multi-stream HMM/ANN system for noise robust speech recognition. Best results were obtained by combining all the classifiers having entropy below average using a weighting proportional to their inverse entropy.


ieee automatic speech recognition and understanding workshop | 2003

Mel-cepstrum modulation spectrum (MCMS) features for robust ASR

Vivek Tyagi; Iain A. McCowan; Hemant Misra

In this paper, we present new dynamic features derived from the modulation spectrum of the cepstral trajectories of the speech signal. Cepstral trajectories are projected over the basis of sines and cosines yielding the cepstral modulation frequency response of the speech signal. We show that the different sines and cosines basis vectors select different modulation frequencies, whereas the frequency responses of the delta and the double delta filters are only centered over 15 Hz. Therefore, projecting cepstral trajectories over the basis of sines and cosines yield a more complementary and discriminative range of features. In this work, the cepstrum reconstructed from the lower cepstral modulation frequency components is used as the static feature. In experiments, it is shown that, as well as providing an improvement in clean conditions, these new dynamic features yield a significant increase in the speech recognition performance in various noise conditions when compared directly to the standard temporal derivative features and C-JRASTA PLP features.


international symposium on neural networks | 1999

Analysis of autoassociative mapping neural networks

Shajith Ikbal; Hemant Misra; B. Yegnanarayana

In this paper we analyse the mapping behavior of an autoassociative neural network (AANN). The mapping in an AANN is achieved by using a dimension reduction followed by a dimension expansion. One of the major results of the analysis is that, the network performs better autoassociation as the size increases. This is because, a network of a given size can deal with only a certain level of nonlinearity. Performance of autoassociative mapping is illustrated with 2D examples. We have shown the utility of the mapping feature of an AANN for speaker verification.


Speech Communication | 2012

Phase AutoCorrelation (PAC) features for noise robust speech recognition

Shajith Ikbal; Hemant Misra; Hynek Hermansky; Mathew Magimai-Doss

In this paper, we introduce a new class of noise robust features derived from an alternative measure of autocorrelation representing the phase variation of speech signal frame over time. These features, referred to as Phase AutoCorrelation (PAC) features include PAC-spectrum and PAC-MFCC, among others. In traditional autocorrelation, correlation between two time delayed signal vectors is computed as their dot product. Whereas in PAC, angle between the vectors in the signal vector space is used to compute the correlation. PAC features are more noise robust because the angle is typically less affected by noise than the dot product. However, the use of angle as correlation estimate makes the PAC features inferior in clean speech. In this paper, we circumvent this problem by introducing another set of features where complementary information among the PAC features and the traditional features are combined adaptively to retain the best of both. An entropy based feature combination method in a multi-layer perceptron (MLP) based multi-stream framework is used to derive an adaptively combined representation of the component feature streams. An evaluation of the combined features using OGI Numbers95 database and Aurora-2 database under various noise conditions and noise levels show significant improvements in recognition accuracies in clean as well as noisy conditions.


international conference on acoustics, speech, and signal processing | 2004

Phase autocorrelation (PAC) features in entropy based multi-stream for robust speech recognition

Shajith Ikbal; Hemant Misra; Hynek Hermansky

Methods to improve noise robustness of speech recognition systems often result in degradation of recognition performance for clean speech. Recently proposed phase autocorrelation (PAC) based features (S. Ikbal et al., Proc. ICASSP-03, p.II-133-6, 2003; Proc. IEEE ASRU 2003 Workshop, 2003), while showing noticeable improvement in noise robustness, also suffer from this drawback. We try to alleviate this problem by using the PAC based features along with regular speech features in a multi-stream framework. The multi-stream system uses the entropy of the posterior probability distribution, computed during recognition, as a confidence measure to combine evidence from different feature streams adaptively (Misra, H. et al., Proc. ICASSP-03, p.II-741-4, 2003). Experimental results obtained on OGI Numbers95 database and Noisex92 noise database show that such a system yields the best possible recognition performance in all conditions. Actually, the combination always performs better than the best performing stream for all the conditions.


Journal of Medical Ultrasound | 2017

Ultrasonography-based Fetal Weight Estimation: Finding an Appropriate Model for an Indian Population

Sujitkumar S. Hiwale; Hemant Misra; Shrutin Ulman

Background: Very limited information is available regarding the accuracy and applicability of various ultrasonography parameters [abdominal circumference (AC), biparietal diameter (BPD), femur length (FL), and head circumference (HC)]-based fetal weight estimation models for Indian population. The objective of this study was to systematically evaluate commonly used fetal weight estimation models to determine their appropriateness for an Indian population. Methods: Retrospective data of 300 pregnant women was collected from a tertiary care center in Bengaluru, India. The inclusion criteria were a live singleton pregnancy, gestational age ≥ 34 weeks, and last ultrasound scan to delivery duration ≤ 7 days. Cases with suspected fetal growth restriction or malformation were excluded. For each case, fetal weight was estimated using 34 different models. The models specifically designed for low birth weight, small for gestation age, or macrosomic babies were excluded. The models were ranked based on their mean percentage error (MPE) and its standard deviation (random error). A model with the least MPE and random error ranking was considered as the best model. Results: In total, 149 cases were found suitable for the study. Out of 34, only 12 models had MPE within ± 10% and only seven models had random error < 10%. Most of the Western population-based models had a tendency to overestimate the fetal weight. Based on MPE and random error ranking, the Woo’s (AC-BPD) model was found to be the best, followed by Jordaan (AC), Combs (AC-HC-FL), Hadlock (AC-HC), and Hadlock-3 (AC-HC-FL) models. It was observed that the models based on just AC and AC-BPD combinations had statistically significant lesser MPE than the models based on all other combinations (p < 0.05). Conclusion: It was observed that the existing models have higher errors on Indian population than on their native populations. This points toward limitations in direct application of these models on Indian population without due consideration. Therefore, it is recommended that clinicians should exert caution in interpretation of fetal weight estimations based on these models. Moreover, this study highlights a need of models based on native Indian population.


Archive | 2013

Topic Modeling for Content Based Image Retrieval

Hemant Misra; Anuj Goyal; Joemon M. Jose

Latent Dirichlet allocation (LDA) topic model has taken a center stage in multimedia information retrieval, for example, LDA model was used by several participants in the recent TRECVid evaluation “Search” task. One of the common approaches while using LDA is to train the model on a set of test images and obtain their topic distribution. During retrieval, the likelihood of a query image is computed given the topic distribution of the test images, and the test images with the highest likelihood are returned as the most relevant images. In this paper we propose to project the unseen query images also in the topic space, and then estimate the similarity between a query image and the test images in the semantic topic space. The positive results obtained by the proposed method indicate that the semantic matching in topic space leads to a better performance than conventional likelihood based approach; there is an improvement of 25 % absolute in the number of relevant results extracted by the proposed LDA based system over the conventional likelihood based LDA system. Another not-so-obvious benefit of the proposed approach is a significant reduction in computational cost.


Ultrasonography | 2018

Fetal weight estimation by ultrasonography: development of Indian population-based models

Sujitkumar S. Hiwale; Hemant Misra; Shrutin Ulman

Purpose Existing ultrasound-based fetal weight estimation models have been shown to have high errors when used in the Indian population. Therefore, the primary objective of this study was to develop Indian population-based models for fetal weight estimation, and the secondary objective was to compare their performance against established models. Methods Retrospectively collected data from 173 cases were used in this study. The inclusion criteria were a live singleton pregnancy and an interval from the ultrasound scan to delivery of ≤7 days. Multiple stepwise regression (MSR) and lasso regression methods were used to derive fetal weight estimation models using a randomly selected training group (n=137) with cross-products of abdominal circumference (AC), biparietal diameter (BPD), head circumference (HC), and femur length (FL) as independent variables. In the validation group (n=36), the bootstrap method was used to compare the performance of the new models against 12 existing models. Results The equations for the best-fit models obtained using the MSR and lasso methods were as follows: log10(EFW)=2.7843700+0.0004197(HC×AC)+0.0008545(AC×FL) and log10(EFW)=2.38 70211110+0.0074323216(HC)+0.0186555940(AC)+0.0013463735(BPD×FL)+0.0004519715 (HC×FL), respectively. In the training group, both models had very low systematic errors of 0.01% (±7.74%) and -0.03% (±7.70%), respectively. In the validation group, the performance of these models was found to be significantly better than that of the existing models. Conclusion The models presented in this study were found to be superior to existing models of ultrasound-based fetal weight estimation in the Indian population. We recommend a thorough evaluation of these models in independent studies.


international conference on acoustics, speech, and signal processing | 2004

Spectral entropy based feature for robust ASR

Hemant Misra; Shajith Ikbal; Hynek Hermansky


international conference on acoustics, speech, and signal processing | 2003

Phase autocorrelation (PAC) derived robust speech features

Shajith Ikbal; Hemant Misra

Collaboration


Dive into the Hemant Misra's collaboration.

Top Co-Authors

Avatar
Top Co-Authors

Avatar

Hervé Bourlard

École Polytechnique Fédérale de Lausanne

View shared research outputs
Top Co-Authors

Avatar

Vivek Tyagi

Idiap Research Institute

View shared research outputs
Top Co-Authors

Avatar

Iain A. McCowan

Queensland University of Technology

View shared research outputs
Top Co-Authors

Avatar

Jithendra Vepa

Idiap Research Institute

View shared research outputs
Top Co-Authors

Avatar
Top Co-Authors

Avatar

B. Yegnanarayana

International Institute of Information Technology

View shared research outputs
Researchain Logo
Decentralizing Knowledge