Pub Date : 2013-12-01DOI: 10.1109/INDCON.2013.6726052
M. S. Patwardhan, W. Gandhare
This study was undertaken to model the tomato seed germination. Tomato seeds were subjected to electrostatic field treatments. It was found that its germination was sensitive to electrostatic field applied. The voltage gradient of 2 kV/mm for 20 second duration was found best with 100% germination in comparison with 92% germination for untreated. The relationship between seed germination and voltage gradient along with time of operation was developed using statistical approach.
{"title":"Enhancement of seed germination by applying electrostatic field","authors":"M. S. Patwardhan, W. Gandhare","doi":"10.1109/INDCON.2013.6726052","DOIUrl":"https://doi.org/10.1109/INDCON.2013.6726052","url":null,"abstract":"This study was undertaken to model the tomato seed germination. Tomato seeds were subjected to electrostatic field treatments. It was found that its germination was sensitive to electrostatic field applied. The voltage gradient of 2 kV/mm for 20 second duration was found best with 100% germination in comparison with 92% germination for untreated. The relationship between seed germination and voltage gradient along with time of operation was developed using statistical approach.","PeriodicalId":313185,"journal":{"name":"2013 Annual IEEE India Conference (INDICON)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126637109","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/INDCON.2013.6726018
Shiburaj Pappu, K. Talele, K. Mehul
Scheduling & Optimization problems are iterative in nature. To find a ideal solution to which is a complex task. These types of problems may be effectively solved and optimal solutions which may be close to the ideal solution may be derived with the help of evolutionary algorithms like the Genetic Algorithm. This paper introduces a new variant of genetic algorithm called Modified Enhanced Steady State Genetic Algorithm (MESSGA) which uses Fuzzy Logic on crossover probability, mutation probability and insertion, for better convergence time. The results of this paper are studied on a common scheduling problem faced by all universities to assign externals for viva-vose or examination to other colleges under its jurisdiction.
{"title":"Modified enhanced steady state genetic algorithm for Scheduling & Optimization problems","authors":"Shiburaj Pappu, K. Talele, K. Mehul","doi":"10.1109/INDCON.2013.6726018","DOIUrl":"https://doi.org/10.1109/INDCON.2013.6726018","url":null,"abstract":"Scheduling & Optimization problems are iterative in nature. To find a ideal solution to which is a complex task. These types of problems may be effectively solved and optimal solutions which may be close to the ideal solution may be derived with the help of evolutionary algorithms like the Genetic Algorithm. This paper introduces a new variant of genetic algorithm called Modified Enhanced Steady State Genetic Algorithm (MESSGA) which uses Fuzzy Logic on crossover probability, mutation probability and insertion, for better convergence time. The results of this paper are studied on a common scheduling problem faced by all universities to assign externals for viva-vose or examination to other colleges under its jurisdiction.","PeriodicalId":313185,"journal":{"name":"2013 Annual IEEE India Conference (INDICON)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126603500","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/INDCON.2013.6726011
C. Paidimarry, B. P. Kumar, S. Katkoori
The performance of modern integrated circuits is significantly influenced by coupling effects of interconnects. Crosstalk noise must therefore be analyzed in the early stage of design to develop reliable VLSI interconnects. In this paper, we present a closed form crosstalk noise estimation technique to coupled RLC interconnects based on coupled transmission line theory and Fourier series analysis. The salient features of our model include accuracy, computational efficiency and continuous time domain solution. We analyze the impact of finite resistance in coupled RLC interconnects. The nonlinear behavior of the driver (CMOS inverter) is captured well in the proposed model. We introduce a novel technique to model CMOS inverter, namely MRLM (Multiple Ramp Library Model). It replaces the nonlinear driver into its equivalent RC model. The resultant circuit resembles with the simple coupled RLC interconnect in the absence of inverters and hence the same analysis is extended after the replacement. Based on our continuous time defined models, we derive analytical models for delay and energy dissipation due to interconnect. In addition, this work emphasizes on improving the performance namely computational efficiency. Simulations of our analytical models and SPICE are demonstrated by assuming different sets of line parameters. It is shown from the numerical results that our estimates match closely with that of SPICE simulations at a mean estimation error of 2.25 percentage.
{"title":"A novel approach to crosstalk noise analysis in CMOS inverter driven coupled RLC interconnects","authors":"C. Paidimarry, B. P. Kumar, S. Katkoori","doi":"10.1109/INDCON.2013.6726011","DOIUrl":"https://doi.org/10.1109/INDCON.2013.6726011","url":null,"abstract":"The performance of modern integrated circuits is significantly influenced by coupling effects of interconnects. Crosstalk noise must therefore be analyzed in the early stage of design to develop reliable VLSI interconnects. In this paper, we present a closed form crosstalk noise estimation technique to coupled RLC interconnects based on coupled transmission line theory and Fourier series analysis. The salient features of our model include accuracy, computational efficiency and continuous time domain solution. We analyze the impact of finite resistance in coupled RLC interconnects. The nonlinear behavior of the driver (CMOS inverter) is captured well in the proposed model. We introduce a novel technique to model CMOS inverter, namely MRLM (Multiple Ramp Library Model). It replaces the nonlinear driver into its equivalent RC model. The resultant circuit resembles with the simple coupled RLC interconnect in the absence of inverters and hence the same analysis is extended after the replacement. Based on our continuous time defined models, we derive analytical models for delay and energy dissipation due to interconnect. In addition, this work emphasizes on improving the performance namely computational efficiency. Simulations of our analytical models and SPICE are demonstrated by assuming different sets of line parameters. It is shown from the numerical results that our estimates match closely with that of SPICE simulations at a mean estimation error of 2.25 percentage.","PeriodicalId":313185,"journal":{"name":"2013 Annual IEEE India Conference (INDICON)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117348935","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/INDCON.2013.6725964
Dinesh Mute, Subhojit Ghosh, B. Subudhi
In this paper an iterative learning controller (ILC) is designed based on the identified model of a single-link flexible manipulator (SLFM). As the system is nonlinear and time-varying so to meet the demands of the control system design, an adaptive nonlinear autoregressive with exogenous input (NARX) model is identified using the input/output experimental data. Tuning of the ILC controller is carried out using least square method. Simulation results demonstrate the potential of the NARX model based ILC controller for precise rotation tracking of a single-link flexible manipulator with suppressing link vibration.
{"title":"Iterative learning control of a single-link flexible manipulator based on an identified adaptive NARX model","authors":"Dinesh Mute, Subhojit Ghosh, B. Subudhi","doi":"10.1109/INDCON.2013.6725964","DOIUrl":"https://doi.org/10.1109/INDCON.2013.6725964","url":null,"abstract":"In this paper an iterative learning controller (ILC) is designed based on the identified model of a single-link flexible manipulator (SLFM). As the system is nonlinear and time-varying so to meet the demands of the control system design, an adaptive nonlinear autoregressive with exogenous input (NARX) model is identified using the input/output experimental data. Tuning of the ILC controller is carried out using least square method. Simulation results demonstrate the potential of the NARX model based ILC controller for precise rotation tracking of a single-link flexible manipulator with suppressing link vibration.","PeriodicalId":313185,"journal":{"name":"2013 Annual IEEE India Conference (INDICON)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128755136","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/INDCON.2013.6725928
K. N. Nischal, M. P. Nayak, K. Manikantan, S. Ramachandran
The appearance of the face varies drastically when background, pose and illumination change. Variations in these conditions make Face Recognition (FR) an even more challenging and difficult task. In this paper, we propose two novel preprocessing techniques, viz., Entropy Augmented Face Isolation (EAFI) and Image Folding, to improve the performance of the FR system. EAFI is used to localize the face region to minimize the effect of cluttered background, thereby enhancing face recognition. Image Folding uses the property of vertical symmetry in the human face to normalize pose variance. The resulting pre-processed image contains the salient details of the face and prepares the ground for feature extraction. Individual stages of the FR system are examined and an attempt is made to improve each stage. DWT and DCT are used for efficient feature extraction and a Binary Particle Swarm Optimization (BPSO) based feature selection algorithm is used to search the feature space for the optimal feature subset. Experimental results show the promising performance of the proposed techniques for face recognition on three benchmark face databases, namely, Color FERET, CMUPIE and Georgia Tech (GT) databases.
{"title":"Face recognition using entropy-augmented face isolation and image folding as pre-processing techniques","authors":"K. N. Nischal, M. P. Nayak, K. Manikantan, S. Ramachandran","doi":"10.1109/INDCON.2013.6725928","DOIUrl":"https://doi.org/10.1109/INDCON.2013.6725928","url":null,"abstract":"The appearance of the face varies drastically when background, pose and illumination change. Variations in these conditions make Face Recognition (FR) an even more challenging and difficult task. In this paper, we propose two novel preprocessing techniques, viz., Entropy Augmented Face Isolation (EAFI) and Image Folding, to improve the performance of the FR system. EAFI is used to localize the face region to minimize the effect of cluttered background, thereby enhancing face recognition. Image Folding uses the property of vertical symmetry in the human face to normalize pose variance. The resulting pre-processed image contains the salient details of the face and prepares the ground for feature extraction. Individual stages of the FR system are examined and an attempt is made to improve each stage. DWT and DCT are used for efficient feature extraction and a Binary Particle Swarm Optimization (BPSO) based feature selection algorithm is used to search the feature space for the optimal feature subset. Experimental results show the promising performance of the proposed techniques for face recognition on three benchmark face databases, namely, Color FERET, CMUPIE and Georgia Tech (GT) databases.","PeriodicalId":313185,"journal":{"name":"2013 Annual IEEE India Conference (INDICON)","volume":"7 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133622815","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/INDCON.2013.6725849
Pradosh Ranjan Sahoo, Subhojit Ghosh
In this paper evolutionary algorithms (EAs) such as genetic algorithm (GA) and differential evolution (DE) optimization are implemented for design of optimal multiloop PI controller with the minimum integral absolute error (IAE) criterion. As a case study multiloop PI control structure for binary distillation column plant described by Wood and Berry, having 2 inputs and outputs has been taken. Simulation is carried out for GA and DE with minimization of IAE and the performances are compared for statistical measures like best, mean and standard deviation of results and, average computation time over 20 independent trials.
{"title":"Evolutionary optimization based tuning for multivarible process","authors":"Pradosh Ranjan Sahoo, Subhojit Ghosh","doi":"10.1109/INDCON.2013.6725849","DOIUrl":"https://doi.org/10.1109/INDCON.2013.6725849","url":null,"abstract":"In this paper evolutionary algorithms (EAs) such as genetic algorithm (GA) and differential evolution (DE) optimization are implemented for design of optimal multiloop PI controller with the minimum integral absolute error (IAE) criterion. As a case study multiloop PI control structure for binary distillation column plant described by Wood and Berry, having 2 inputs and outputs has been taken. Simulation is carried out for GA and DE with minimization of IAE and the performances are compared for statistical measures like best, mean and standard deviation of results and, average computation time over 20 independent trials.","PeriodicalId":313185,"journal":{"name":"2013 Annual IEEE India Conference (INDICON)","volume":"183 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127022221","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/INDCON.2013.6725886
D. Mahapatra, H. Chanakya, T. Ramachandra
Influx of sewage into surface water results in nutrient enrichment and consequently leads to algal bloom and voluminous organic sludge production in urban areas as in case of Bangalore. The lack of utilities of algae and resulting sludge has lead to anoxia and GHG emissions. Environmental friendly ways of sludge and algal biomass utilities as well as disposal are scant. Testing for biofuel and thermal properties can be beneficial to meet the energy requirement to run treatment plants that might have better fuel value to increase the net energy gain in the system. As the volume of algal biomass and quantity of sludge has increased over the past few years, sustainable means of biomass and sludge utilization needs to be devised for beneficial purposes, to keep the surface waters clean and regulate the biomass productivity of such systems. Therefore the biomass and sludge characterization becomes imperative for any further utilities. In the present study the indigenous suspended algae and the algal bioreactor sludge were characterized for the different functionalities and presence of bio-chemicals (carbohydrates, proteins and lipids) through Infrared Analysis (ATR-FTIR). The total lipids and fatty acid methyl esters (FAME) composition were studied. The heat values and thermal decomposition pattern were analyzed by [Thermogravimetry (TGA)/Differential Thermal Analysis (DT) and Differential Scanning Calorimetry (DSC)]. The algae were found to have a better total lipid content of >34 % compared to 22% in case of wastewater sludge with quality FAME for better biofuel properties. However there were higher number of FAME in wastewater sludge (>30) where C16 and C18 members dominated. Algal biomass showed higher calorific value of 17.96 MJkg-1 compared to 10.33 MJkg-1 of wastewater sludge.
{"title":"Bioenergy generation from components of a Continuous algal bioreactor: Analysis of lipids, spectroscopic and thermal properties","authors":"D. Mahapatra, H. Chanakya, T. Ramachandra","doi":"10.1109/INDCON.2013.6725886","DOIUrl":"https://doi.org/10.1109/INDCON.2013.6725886","url":null,"abstract":"Influx of sewage into surface water results in nutrient enrichment and consequently leads to algal bloom and voluminous organic sludge production in urban areas as in case of Bangalore. The lack of utilities of algae and resulting sludge has lead to anoxia and GHG emissions. Environmental friendly ways of sludge and algal biomass utilities as well as disposal are scant. Testing for biofuel and thermal properties can be beneficial to meet the energy requirement to run treatment plants that might have better fuel value to increase the net energy gain in the system. As the volume of algal biomass and quantity of sludge has increased over the past few years, sustainable means of biomass and sludge utilization needs to be devised for beneficial purposes, to keep the surface waters clean and regulate the biomass productivity of such systems. Therefore the biomass and sludge characterization becomes imperative for any further utilities. In the present study the indigenous suspended algae and the algal bioreactor sludge were characterized for the different functionalities and presence of bio-chemicals (carbohydrates, proteins and lipids) through Infrared Analysis (ATR-FTIR). The total lipids and fatty acid methyl esters (FAME) composition were studied. The heat values and thermal decomposition pattern were analyzed by [Thermogravimetry (TGA)/Differential Thermal Analysis (DT) and Differential Scanning Calorimetry (DSC)]. The algae were found to have a better total lipid content of >34 % compared to 22% in case of wastewater sludge with quality FAME for better biofuel properties. However there were higher number of FAME in wastewater sludge (>30) where C16 and C18 members dominated. Algal biomass showed higher calorific value of 17.96 MJkg-1 compared to 10.33 MJkg-1 of wastewater sludge.","PeriodicalId":313185,"journal":{"name":"2013 Annual IEEE India Conference (INDICON)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130717238","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/INDCON.2013.6726117
Karam Singh, S. Ahamed
Motion estimation (ME) is one of the most important and complex block of all the existing video coding standards. In H.264/AVC video coding standard ME process is much more complex due to variable block size with quarter pixel accuracy. In this paper, we proposed a modified small-cross diamond search (MSCDS) motion estimation algorithm for H.264/AVC video coding standard. The proposed algorithm reduces the number of checking points and thereby makes the motion estimation process faster. Moreover, the MSCDS algorithm is suitable for stationary, quasi-stationary and fast moving video sequences and computationally less complex from hardware point of view. Simulation results show that the proposed algorithm requires very few number of search points for finding the best matched block with almost negligible loss in video quality. As compared to the fastest existing ME algorithm, proposed algorithm requires 9.02% less number of search points with 0.007 dB loss in PSNR.
{"title":"Modified small-cross diamond search motion estimation algorithm for H.264/AVC","authors":"Karam Singh, S. Ahamed","doi":"10.1109/INDCON.2013.6726117","DOIUrl":"https://doi.org/10.1109/INDCON.2013.6726117","url":null,"abstract":"Motion estimation (ME) is one of the most important and complex block of all the existing video coding standards. In H.264/AVC video coding standard ME process is much more complex due to variable block size with quarter pixel accuracy. In this paper, we proposed a modified small-cross diamond search (MSCDS) motion estimation algorithm for H.264/AVC video coding standard. The proposed algorithm reduces the number of checking points and thereby makes the motion estimation process faster. Moreover, the MSCDS algorithm is suitable for stationary, quasi-stationary and fast moving video sequences and computationally less complex from hardware point of view. Simulation results show that the proposed algorithm requires very few number of search points for finding the best matched block with almost negligible loss in video quality. As compared to the fastest existing ME algorithm, proposed algorithm requires 9.02% less number of search points with 0.007 dB loss in PSNR.","PeriodicalId":313185,"journal":{"name":"2013 Annual IEEE India Conference (INDICON)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131281186","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/INDCON.2013.6726073
Vijay Kumar Dixit, Rajeev Gupta, K. Pal
This paper presents three voltage-mode first order all pass filters using differential difference current conveyors (DDCC) as the active elements. The first circuit has one DDCC, one capacitor and one grounded resistor. The second circuit also employs one DDCC, one resistor and one grounded capacitor. The third circuit uses two DDCC, one resistor and one capacitor and both the elements are grounded. These circuits have the following advantages: high input impedance, low passive sensitivities and no requirements for component matching conditions and in the third circuit all passive components are grounded and thus it is preferred for integrated circuit implementation. Theoretical results are verified using PSPICE simulation.
{"title":"High-input impedance first order all-pass filters using DDCC","authors":"Vijay Kumar Dixit, Rajeev Gupta, K. Pal","doi":"10.1109/INDCON.2013.6726073","DOIUrl":"https://doi.org/10.1109/INDCON.2013.6726073","url":null,"abstract":"This paper presents three voltage-mode first order all pass filters using differential difference current conveyors (DDCC) as the active elements. The first circuit has one DDCC, one capacitor and one grounded resistor. The second circuit also employs one DDCC, one resistor and one grounded capacitor. The third circuit uses two DDCC, one resistor and one capacitor and both the elements are grounded. These circuits have the following advantages: high input impedance, low passive sensitivities and no requirements for component matching conditions and in the third circuit all passive components are grounded and thus it is preferred for integrated circuit implementation. Theoretical results are verified using PSPICE simulation.","PeriodicalId":313185,"journal":{"name":"2013 Annual IEEE India Conference (INDICON)","volume":"118 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132220563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/INDCON.2013.6726131
Sourjya Sarkar, K. S. Rao, Dipanjan Nandi, S. B. S. Kumar
In this paper we explore the performance of multilingual speaker recognition systems developed on the IITKGP-MLILSC speech corpus. Closed-set speaker identification and speaker verification experiments are individually conducted on 13 widely spoken Indian languages. In particular, we focus on the effect of language mismatch in the speaker recognition performance of individual languages and all languages together. The standard GMM-based speaker recognition framework is used. While the average language-independent speaker identification rate is as high as 95.21%, an average equal error rate of 11.71% shows scope for further improvement in speaker verification performance.
{"title":"Multilingual speaker recognition on Indian languages","authors":"Sourjya Sarkar, K. S. Rao, Dipanjan Nandi, S. B. S. Kumar","doi":"10.1109/INDCON.2013.6726131","DOIUrl":"https://doi.org/10.1109/INDCON.2013.6726131","url":null,"abstract":"In this paper we explore the performance of multilingual speaker recognition systems developed on the IITKGP-MLILSC speech corpus. Closed-set speaker identification and speaker verification experiments are individually conducted on 13 widely spoken Indian languages. In particular, we focus on the effect of language mismatch in the speaker recognition performance of individual languages and all languages together. The standard GMM-based speaker recognition framework is used. While the average language-independent speaker identification rate is as high as 95.21%, an average equal error rate of 11.71% shows scope for further improvement in speaker verification performance.","PeriodicalId":313185,"journal":{"name":"2013 Annual IEEE India Conference (INDICON)","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133160102","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：