Pub Date : 1994-04-19DOI: 10.1109/ICASSP.1994.389355
A. Peinado, J. C. Segura, A. Rubio, M. C. Benítez
Although the continuous HMM (CHMM) technique seems to be the most flexible and complete tool for speech modeling, it is not always used for the implementation of speech recognition systems due to several problems related to training and computational complexity. Besides, it is not clear the superiority of continuous models over other well-known types of HMMs, such as discrete (DHMM) or semicontinuous (SCHMM) models, or multiple vector quantization (MVQ) models, a new type of HMM modeling. The authors propose a new variant of HMM models, the SCMVQ, HMM models (semicontinuous multiple vector quantization HMM), that uses one VQ codebook per recognition unit and several quantization candidates, Formally, SCMVQ modeling is the closest one to CHMM, although requiring less computation than SCHMMs. Besides, the authors show that SCMVQs can obtain better recognition results than DHMMs, SCHMMs or MVQs.<>
{"title":"Using multiple vector quantization and semicontinuous hidden Markov models for speech recognition","authors":"A. Peinado, J. C. Segura, A. Rubio, M. C. Benítez","doi":"10.1109/ICASSP.1994.389355","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389355","url":null,"abstract":"Although the continuous HMM (CHMM) technique seems to be the most flexible and complete tool for speech modeling, it is not always used for the implementation of speech recognition systems due to several problems related to training and computational complexity. Besides, it is not clear the superiority of continuous models over other well-known types of HMMs, such as discrete (DHMM) or semicontinuous (SCHMM) models, or multiple vector quantization (MVQ) models, a new type of HMM modeling. The authors propose a new variant of HMM models, the SCMVQ, HMM models (semicontinuous multiple vector quantization HMM), that uses one VQ codebook per recognition unit and several quantization candidates, Formally, SCMVQ modeling is the closest one to CHMM, although requiring less computation than SCHMMs. Besides, the authors show that SCMVQs can obtain better recognition results than DHMMs, SCHMMs or MVQs.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133852969","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389459
Wentao Zheng, H. Harashima
This paper presents an algorithmic procedure for the generation of a 3D wireframe model from range image. The basic idea is to represent a surface by dominant points which have important shape attributes. To determine dominant points, we propose a criterion based on 3D invariant characteristics of surfaces, which we call mean-square curvature. This quantity has some desirable properties and, compared to conventional Gaussian curvature or mean curvature, is more suitable for dominant point selection. It also allows a physical explanation. The 3D wireframe model is constructed by detecting dominant points followed by triangulating them in 3D space. The algorithms are tested on a real range image and the results are shown.<>
{"title":"The automatic generation of 3D object model from range image","authors":"Wentao Zheng, H. Harashima","doi":"10.1109/ICASSP.1994.389459","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389459","url":null,"abstract":"This paper presents an algorithmic procedure for the generation of a 3D wireframe model from range image. The basic idea is to represent a surface by dominant points which have important shape attributes. To determine dominant points, we propose a criterion based on 3D invariant characteristics of surfaces, which we call mean-square curvature. This quantity has some desirable properties and, compared to conventional Gaussian curvature or mean curvature, is more suitable for dominant point selection. It also allows a physical explanation. The 3D wireframe model is constructed by detecting dominant points followed by triangulating them in 3D space. The algorithms are tested on a real range image and the results are shown.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124114663","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389789
T. Durrani, A. R. Leyman, J. Soraghan
This paper brings together two strands of current interest in signal processing. While second order techniques and minimum variance criteria are well understood, there is a growing requirement to study the performance criteria that involve higher order statistics in order to evaluate deviation from Gaussianity, linearity and stationarity of observed data. There is a complimentary requirement for the generation of random test sequences which have prescribed (or minimal) higher order statistics, to facilitate the analysis of systems in order to determine linearity/non-linearity, time invariance vs time varying parameters. This paper proposes a new method for minimising the third order cumulant spread of random sequences with symmetric pdf, and provides a closed form solution for the weightings required to achieve this. Numerous computed results are included to verify performance.<>
{"title":"\"Whiter than white\" noise","authors":"T. Durrani, A. R. Leyman, J. Soraghan","doi":"10.1109/ICASSP.1994.389789","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389789","url":null,"abstract":"This paper brings together two strands of current interest in signal processing. While second order techniques and minimum variance criteria are well understood, there is a growing requirement to study the performance criteria that involve higher order statistics in order to evaluate deviation from Gaussianity, linearity and stationarity of observed data. There is a complimentary requirement for the generation of random test sequences which have prescribed (or minimal) higher order statistics, to facilitate the analysis of systems in order to determine linearity/non-linearity, time invariance vs time varying parameters. This paper proposes a new method for minimising the third order cumulant spread of random sequences with symmetric pdf, and provides a closed form solution for the weightings required to achieve this. Numerous computed results are included to verify performance.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124117717","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389287
C. Weinstein
The ARPA Spoken Language Technology and Applications Day (SLTA'93) was a special workshop which presented a set of live, state-of-the-art demonstrations of speech recognition and spoken language understanding systems. The purpose of this paper is to provide perspective on current opportunities for applications of spoken language technology by summarizing the demonstrations and the related applications which they can enable, and reviewing the applications opportunities and needs cited by panelists and other members of the user community.<>
{"title":"Demonstrations and applications of spoken language technology: highlights and perspectives from the 1993 ARPA Spoken Language Technology and Applications Day","authors":"C. Weinstein","doi":"10.1109/ICASSP.1994.389287","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389287","url":null,"abstract":"The ARPA Spoken Language Technology and Applications Day (SLTA'93) was a special workshop which presented a set of live, state-of-the-art demonstrations of speech recognition and spoken language understanding systems. The purpose of this paper is to provide perspective on current opportunities for applications of spoken language technology by summarizing the demonstrations and the related applications which they can enable, and reviewing the applications opportunities and needs cited by panelists and other members of the user community.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"144 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124397026","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389598
M. Ibnkahla, S. Puechmorel, F. Castanie
Many signal processing problems need to be solved in an adaptive way under some constraints. The paper introduces a constrained complex-valued neural network (CCNN) model. It is composed of two sub networks: a master which gives the main energy function (the error power between the master's output and a desired output), and a slave which gives a secondary energy function (related to the constraints imposed by the problem). The sum of these energy functions gives the cost function to be minimized by the CCNN. An extension of the classical back propagation algorithm to the complex plane, under some inequality constraints, is used for the training process. This model finds a natural application in the time-frequency analysis as it gives direct access to the time-frequency signature.<>
{"title":"A constrained neural network with complex activation function: application to time-frequency analysis","authors":"M. Ibnkahla, S. Puechmorel, F. Castanie","doi":"10.1109/ICASSP.1994.389598","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389598","url":null,"abstract":"Many signal processing problems need to be solved in an adaptive way under some constraints. The paper introduces a constrained complex-valued neural network (CCNN) model. It is composed of two sub networks: a master which gives the main energy function (the error power between the master's output and a desired output), and a slave which gives a secondary energy function (related to the constraints imposed by the problem). The sum of these energy functions gives the cost function to be minimized by the CCNN. An extension of the classical back propagation algorithm to the complex plane, under some inequality constraints, is used for the training process. This model finds a natural application in the time-frequency analysis as it gives direct access to the time-frequency signature.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"194 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114304191","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389900
D. Maiwald, J. Böhme
Investigates the estimation of both the number of waves and of the wave parameters for wavefields in a geophysical application. A parametric method for wave parameter estimation in connection with a multiple test procedure is presented. The distribution of the corresponding test statistic for wideband data is approximated by the central limit theorem and alternatively by a bootstrap procedure. Finally the application of the algorithms to real seismic data is studied.<>
{"title":"Multiple testing for seismic data using bootstrap","authors":"D. Maiwald, J. Böhme","doi":"10.1109/ICASSP.1994.389900","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389900","url":null,"abstract":"Investigates the estimation of both the number of waves and of the wave parameters for wavefields in a geophysical application. A parametric method for wave parameter estimation in connection with a multiple test procedure is presented. The distribution of the corresponding test statistic for wideband data is approximated by the central limit theorem and alternatively by a bootstrap procedure. Finally the application of the algorithms to real seismic data is studied.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114439879","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.390010
L. Resende, J. Romano, M. Bellanger
A robust approach to implement the FLS algorithm for linearly constrained adaptive filtering is derived in this work. The robustness is provided by means of an additional correcting term which is also updated by a LS procedure. In fact, the novel algorithm works as the LS version of the classical LMS-based Frost algorithm. Simulation results with a long data input sequence show the performance of the proposed technique.<>
{"title":"A robust FLS algorithm for linearly-constrained adaptive filtering","authors":"L. Resende, J. Romano, M. Bellanger","doi":"10.1109/ICASSP.1994.390010","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.390010","url":null,"abstract":"A robust approach to implement the FLS algorithm for linearly constrained adaptive filtering is derived in this work. The robustness is provided by means of an additional correcting term which is also updated by a LS procedure. In fact, the novel algorithm works as the LS version of the classical LMS-based Frost algorithm. Simulation results with a long data input sequence show the performance of the proposed technique.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114557478","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389563
R. Sankar, Shrenik Patravali
The multilayer perceptron (MLP) type of neural network classifiers using backpropagation has become increasingly popular for speech recognition. However, for the case of noisy speech, studies have not been very extensive. In this paper, a robust speech recognition system using a neural network is studied. Robustness is achieved by noise immunization, thereby enabling the system to maintain a high recognition accuracy for speech input at different signal-to-noise ratio (SNR) conditions. Noise immunization is achieved by gradual contamination of the signal with noise thereby creating a more reliable reference database in spite of low SNR. The learning is done by a modified backpropagation algorithm. Tenth order LPC coefficients are used to represent the data. The order or sequence in which the data is presented to the neural network for training to provide fast convergence and better performance is studied.<>
{"title":"Noise immunization using neural net for speech recognition","authors":"R. Sankar, Shrenik Patravali","doi":"10.1109/ICASSP.1994.389563","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389563","url":null,"abstract":"The multilayer perceptron (MLP) type of neural network classifiers using backpropagation has become increasingly popular for speech recognition. However, for the case of noisy speech, studies have not been very extensive. In this paper, a robust speech recognition system using a neural network is studied. Robustness is achieved by noise immunization, thereby enabling the system to maintain a high recognition accuracy for speech input at different signal-to-noise ratio (SNR) conditions. Noise immunization is achieved by gradual contamination of the signal with noise thereby creating a more reliable reference database in spite of low SNR. The learning is done by a modified backpropagation algorithm. Tenth order LPC coefficients are used to represent the data. The order or sequence in which the data is presented to the neural network for training to provide fast convergence and better performance is studied.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"180 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114531068","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.390044
Li-Chien Lin, C.-C. Jay Kuo
A new approach for signal extrapolation based on wavelet representation: known as scale-time limited extrapolation and a denoising process is investigated in this research. We first examine a new signal modeling technique using wavelets and the corresponding scale-time limited signal extrapolation algorithm. Then, the sensitivity of the algorithm to noise is discussed, and a denoising algorithm based on the time-localization property of the wavelet transform is proposed. By integrating the denoising process and the iterative scale-time limited extrapolation algorithm, we obtain a very robust signal extrapolation algorithm for noisy data. A simulation result of signal extrapolation from noisy observed data is presented to illustrate the performance of the proposed robust signal extrapolation algorithm.<>
{"title":"Robust signal extrapolation using wavelets","authors":"Li-Chien Lin, C.-C. Jay Kuo","doi":"10.1109/ICASSP.1994.390044","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.390044","url":null,"abstract":"A new approach for signal extrapolation based on wavelet representation: known as scale-time limited extrapolation and a denoising process is investigated in this research. We first examine a new signal modeling technique using wavelets and the corresponding scale-time limited signal extrapolation algorithm. Then, the sensitivity of the algorithm to noise is discussed, and a denoising algorithm based on the time-localization property of the wavelet transform is proposed. By integrating the denoising process and the iterative scale-time limited extrapolation algorithm, we obtain a very robust signal extrapolation algorithm for noisy data. A simulation result of signal extrapolation from noisy observed data is presented to illustrate the performance of the proposed robust signal extrapolation algorithm.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115040271","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389773
Hui Liu, Fu Li
Presents a new approach to simultaneously estimate stacking velocity and zero-offset time of seismic wave propagation, which is specially designed for multiple seismic wavefronts while the traditional semblance approach and the subspace approach are not. Simulations show a good performance of the new approach.<>
{"title":"Maximum likelihood velocity estimation of multiple seismic wavefronts","authors":"Hui Liu, Fu Li","doi":"10.1109/ICASSP.1994.389773","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389773","url":null,"abstract":"Presents a new approach to simultaneously estimate stacking velocity and zero-offset time of seismic wave propagation, which is specially designed for multiple seismic wavefronts while the traditional semblance approach and the subspace approach are not. Simulations show a good performance of the new approach.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115091946","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：