Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
6 results
Search Results
Item Factor analysis methods for joint speaker verification and spoof detection(Institute of Electrical and Electronics Engineers Inc., 2017) Dhanush, B.K.; Suparna, S.; Aarthy, R.; Likhita, C.; Shashank, D.; Harish, H.; Ganapathy, S.The performance of a speaker verification system is severely degraded by spoofing attacks generated from artificial speech synthesizers. Recently, several approaches have been proposed for classifying natural and synthetic speech (spoof detection) which can be used in conjunction with a speaker verification system. In this paper, we attempt to develop a joint modelling approach which can detect the presence of spoofing attacks while also performing the speaker verification task. We propose a factor modelling approach where the spoof variability subspace and the speaker variability subspace are jointly trained. The lower dimensional projections in these subspaces are used for speaker verification as well as spoof detection tasks. We also investigate the benefits of linear discriminant analysis (LDA), widely used in speaker recognition, for the spoof detection task. Several experiments are performed using the speaker and spoofing (SAS) database. For speaker verification, we compare the performance of the proposed method with a baseline method of fusing a conventional speaker verification system and a spoof detection system. In these experiments, the proposed approach provides substantial improvements for spoof detection (relative improvements of 20% in EER over the baseline) as well as speaker verification under spoofing conditions (relative improvements of 40% in EER over the baseline). © 2017 IEEE.Item Shortcomings of Vibrating Screen and Corrective Measures: A Review(Springer Nature, 2020) Shanmugam, B.K.; Vardhan, H.; Govinda Raj, M.; Kaza, M.; Sah, R.; Harish, H.Screening is a process of separating two or more materials of size ranging from fine to coarse of different shapes, particle sizes and densities. The conventional vibrating screen is widely used in mineral and mining industries for performing sizing operation. This paper will be on the review of the various shortcomings of the conventional vibrating screen. The review was carried out through literature survey and plant visit. The paper also involves the remedial measures to be taken to overcome the shortcomings of the conventional vibrating screen. Some of the corrective measures are reduction in number of components also reduces overall screen load, angular velocities, stress, wear or damages to the screen, screen replacement, downtime and overall cost of production. The circular vibrating motion provided to the screen will give larger amplitude and stroke length of the screen which increases screening efficiency. The overall outcome of remedial action will lead to improved screening efficiency. This paper also provides the idea for the optimization of the vibrating screen design which can reduce the power consumption, friction and also provide high screening output. © 2020, Springer Nature Switzerland AG.Item Enhanced Recovery of Iron Values from Low-Grade Ores and Tailings through Reverse Cationic Flotation(World Researchers Associations, 2025) Varma, R.M.; Harsha, V.; Reddy, B.R.R.; Shanmugam, S.B.; Harish, H.India is well-known for its rich deposits of high-quality hematite ores, making it a vital player in the global market. As the availability of high-grade iron ores diminishes, the need to process low-grade ores, fines and slimes through beneficiation is becoming increasingly important to meet market requirements. The creation of fines and slimes leads to a mineral loss of about 20 to 25% of the overall mineral value during processing. This research investigates the beneficiation of iron ore tailings using reverse cationic flotation, with Sokem reagent acting as a collector and starch serving as a depressant. A series of comparative assessments involving magnetic separation and gravity separation were performed. An initial mineralogical examination showed that hematite and goethite were the main iron-bearing minerals, accompanied by quartz and kaolinite as significant gangue materials. The selective flocculation technique proved effective, enhancing the iron grade from 41.05% to 57.03% Fe, with a recovery rate of 47.35%. After desliming, the outcomes improved further, yielding 58.25% Fe and a recovery of 29.00%. These results underline the potential for successful beneficiation of iron ore tailings, offering valuable insights for enhancing the recovery of high-grade iron from low-grade ores and reducing mineral losses during processing. © 2025, World Researchers Associations. All rights reserved.Item Investigation of Dust Emission in Limestone Mines and its Statistical Prediction using Supervised Machine Learning (Regression) Modelling(World Researchers Associations, 2025) Rajib, P.; Harsha, V.; Shanmugam, S.B.; Harish, H.; Amrites, S.In India, the fugitive dust emissions in the processing plant and mining area of limestone mines are very high. The dust emission of (particulate matter) PM10 and PM2.5 forms an unsafe working environment for workers in processing plant areas and mining areas. The excessive emission of PM10 and PM2.5 will cause lung-related diseases to the workers and the people existing in the adjacent areas of the mine. The dust emission majorly causes air pollution to occur due to the distribution of particulate matter in the work area. This study majorly investigates the dust emission levels of PM10 and PM2.5 in the limestone mine of Kadapa, Andra Prasad, India. The investigation on the dust emission of PM10 and PM2.5 was carried out as per the guidelines of DGMS and MoEF and CC guidelines, with a specific focus on PM10 and PM2.5 particulate matter. From the study, it was clear that the dust emission levels of PM10 and PM2.5 in the mine area and some parts of the processing area were below the permissible limit of 1200 ?g/m³ as per the National Ambient Air Quality Standards (NAAQS, 2009). It was also found that the dust emission levels of PM10 and PM2.5 in the crushing and screening area of the processing plant were above the permissible limit of 1200 ?g/m³. Further the statistical prediction model was developed using linear, quadratic and cubic supervised machine learning (regression) modelling. The results indicated that the cubic regression model will provide the accurate prediction of fugitive dust emission with lower error and standard deviation. © 2025, World Researchers Associations. All rights reserved.Item Development of beneficiation circuit for low-grade laterite iron ores sourced from the Gujarat area(World Researchers Associations, 2025) Reddy, B.R.R.; Harsha, V.; Bhushan, A.S.; Harish, H.; Shanmugam, S.B.This study focuses on the maximum recovery of iron values from the low-grade laterite iron ore. The Fe analysis of laterite was carried out using wet method analysis. Subsequently, the characterization studies were carried out on laterite ore using Optical microscope for liberation studies, mineral phase analysis with XRD and elemental analysis using SEM-EDS. Further, the ore of feed particle size of-150 microns was subjected to physical separation techniques such as scrubbing, hydro cyclone, spiral concentrator and dual-stage HGMS and two beneficiation circuits. The results from the above physical separation beneficiation techniques showed a concentrate of 41.25% FeG and a recovery of 48.05% in beneficiation circuit 1 and a concentrate of 48.03 % FeG and a recovery of 62.11% in beneficiation circuit 2 which is not feasible for iron-making in the blast furnace. © 2025, World Researchers Associations. All rights reserved.Item Experimental and statistical analysis on rate of penetration under the influence of rotational speed for drilling limestone in the open cast mine area(World Researchers Associations, 2025) Subrahmanyam, S.K.V.; Harsha, V.; Reddy, B.R.R.; Shanmugam, S.B.; Harish, H.In this study, an experimental investigation was carried out to study the rate of penetration for drilling limestone in an open-cast mine. The investigation was also carried out to study the influence of rotational speed. Drilling experiments were carried out with a constant drilling depth of 10m and varying speeds of 40rpm, 45rpm and 50rpm. As the drilling was carried out, the fresh drill bit caused an increase in the rate of drilling penetration. Further, as it reached the optimal level, there was a decrease in the rate of penetration due to the wearing out of the drill bit. Further, the prediction of experimental results was carried out using the regression analysis using linear and polynomial models. The results show that the polynomial model was found to be in close relation with experimental results. © 2025, World Researchers Associations. All rights reserved.