Cole, R., et al. (1995). The Challenge of Spoken Language Systems: Research Directions for the Nineties. IEEE Transactions on Speech and Audio Processing, 3(1):1-21.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....with respect to other compensation methods reported in the bibliography and reveals the importance of the non linear effects of the noise and the utility of the proposed method. 1. INTRODUCTION The noise severely affects automatic speech recognition applications working in real conditions [1, 2]. The recognition systems, usually trained with clean speech do not model properly the speech acquired under noisy conditions. The noise significantly degrades the performance of speech recognizers mainly due to the mismatch between the training conditions and recognition conditions [3] The ....

R. Cole et. al. The challenge of spoken language systems: research directions for the nineties. IEEE Trans. on Speech and Audio Processing, 3(1):1--21, January 1995.

.... There is currently great interest in increasing the robustness of automatic speech recognition (ASR) to make it more effective in adverse environments e.g. when interfering noise, reverberation, distortion or filtering of the signal is present etc for a review of work in this area, see [4]. However, these techniques are ultimately limited by the amount of information available in the degraded audio signal and there has recently been interest in augmenting the audio signal with a visual signal derived from an image of the speaker s lips (speechreading) 9] At present, it is not ....

R. Cole et al. The challenge of spoken language systems: research directions for the nineties. IEEE Transactions on Speech and Audio Processing, 3(1):1--21, January 1995.

....of such systems will require advance empirical work with human subjects, building a variety of new prototype systems, and the development of appropriate metrics for evaluating the accuracy, efficiency, learnability, expressive power, and other characteristics of different multimodal systems. (Cole, Hirschman et al. 1995, 12) Development and availability of a spoken language enhancement to an interface for the World Wide Web would also increase the availability and visibility of spoken language technology in the computing community as a whole. This may encourage other researchers and developers to refine and ....

....based on physical pointing did not make use of the full range of expressive capabilities of human users. This omission was, no doubt, mostly a consequence of the relatively poor state of other means of expression as input modalities; spoken language systems have made immense progress since 1983 (Cole, Hirschman et al. 1995). Adding spoken language capabilities to hypermedia holds the promise of extending users abilities in ways they find appealing. Empirical studies of multimodal interfaces have looked at user preferences for different kinds of inputs. For example, Rudnicky (1993) showed that users preferred speech ....

Cole, R., Hirschman, L., et al. (1995). The challenge of spoken language systems: Research directions for the Nineties. IEEE Transactions on Speech and Audio Processing, 3(1), p. 1-20.

.... task to be performed by a computer system because the variability in the way people speak, which is reflected in tremendously complex speech signals to be processed in the automatic speech recognition systems (ASRS) There are several key areas of future research which have been pointed out in [3] as significant for the future development of the spoken language systems. These include robust speech recognition; automatic training and adaptation; spontaneous speech; dialogue models; natural language response generation; speech synthesis and speech generation; multi lingual systems; ....

Cole, R. et al (1995) The Challenge of Spoken Language Systems: Research Directions for the Nineties, IEEE Transactions on Speech and Audio Processing, vol.3, No.1, January 1995, 1-21.

....replace anecdotal arguments with scientific evidence [Shneiderman 1993] Bradford [1995] states that there are almost certainly applications where speech is the more natural medium and calls for comparative studies to determine where and when speech functions most effectively as a user interface. Cole et al. 1995] note the role that spoken language should ultimately play in multimodal systems is not well understood and call for the development of theoretical models from which predictions can be made about the strengths, weaknesses, and overall performance of different types of unimodal and multimodal ....

Cole, R., et al. (1995). The Challenge of Spoken Language Systems: Research Directions for the Nineties. IEEE Transactions on Speech and Audio Processing, 3(1):1-21.

....in Multimodal Speech Recognition Environments Page 2 Michael A. Grasso To understand how to leverage this advantage, these anecdotal arguments need to be tested with a scientific approach. More theoretical work is needed in order to help predict the performance of speech in multimodal environments [3, 4, 5]. The focus of this paper, therefore, is to propose a framework to empirically evaluate the types of tasks that might benefit from a multimodal interface. Before exploring this issue, an overview of speech recognition technology is given. This is followed by theoretical work in task integration. ....

Cole, R., et al. The Challenge of Spoken Language Systems: Research Directions for the Nineties. IEEE Transactions on Speech and Audio Processing, 3, 1, (January 1995), pp. 1-21.

....information to help classification. The purpose of this paper is to introduce a two dimensional hidden Markov model (2 D HMM) as a general framework to build context dependent classifiers. Hidden Markov models have earned their popularity mostly from successful application to speech recognition [4, 5, 6]. Despite the weakness of the Markovian assumption as applied to speech, they have proven to be a powerful method in speech processing. The probability mechanism is as follow: at any discrete unit of time, the system is assumed to exist in one of a finite set of states. Within each state there is ....

R. Cole, L. Hirschman, L. Atlas, et al., "The challenge of spoken language systems: research directions for the nineties," IEEE Transactions on Speech and Audio Processing, volume 3, pages 1-21, 1063-6676, Jan. 1995.

....a video signal based on who is in the scene and what they are doing. Such abilities in a computer are hard to imagine, unless it has an ability to perceive people. Speech perception has made much progress in the recent years and some amazing results in word spotting have recently been presented [28, 26, 75]. The computer vision field has also taken to this problem has made some amazing progress in the recent years. The important problems that computer vision aims to address in order to make machines that see people is that it needs to first determine if someone is near it (where) and how many people ....

R. A. Cole, L. Hirschman, and et al. The challenge of spoken language systems: Research directions for the nineties. IEEE Transactions on Speech and Audio Processing, 3(1):1--21, January 1995.

....levels of noise which often hinders the use of speech recognition systems. Much research effort has therefore been directed to systems for noisy environments [1 ] and the robustness of speech recognition systems has been identified as one of the biggest obstacles to overcome in future research [2 ] Most approaches for robust recognition make use of the acoustic speech signal only and ignore the multimodal nature of human speech. Psychological studies have shown that besides the acoustic signal, visual information of the speaker s face is often involved in the recognition process [3 ] ....

....weights thus describe the shape of the model and are used as input features for the recognition system. This method enables detailed shape description with a small number of parameters. The parameters are linearly independent although non linear dependencies might be present between the modes. 2. 2 Locating and Tracking Lips Locating and tracking lips in image sequences is a difficult object recognition problem due to the lack of dominant image features representing the lip contours. The contrast at the outer lip contour is often very small and the contrast at the inner lip contour is ....

[Article contains additional citation context not shown here]

R. Cole, L. Hirschman, L. Atlas et al., "The Challenge of Spoken Language Systems: Research Directions for the Nineties", IEEE Trans. on Speech and Audio Processing, Vol. 3, No. 1, 1995.

....prove indispensable in situations where the operator s hands are occupied such as when driving a car or operating machinery. Much research has focused on the development of spoken language systems and rapid advances in the field of automatic speech recognition (ASR) have been made in recent years [7, 23]. Although progress has been impressive, researchers have yet to overcome the inherent limitations of purely acoustic based systems, particularly their susceptibility to environmental noise. Such systems readily degrade when exposed to non stationary or unpredictable noise as might be encountered ....

R. Cole, L. Hirschmann, L. Atlas, et al. The challenge of spoken language systems: Research directions for the nineties. IEEE Trans. on Speech and Audio Processing, 3(1):1--20, 1995.

....as a general framework for context dependent classifiers. The theory of hidden Markov models in one dimension (1 D HMMs) was developed in the 1960s by Baum, Eagon, Petrie, Soules, and Weiss [5, 6, 7, 8] HMMs have earned their popularity mostly from successful application to speech recognition [9, 10, 11, 12, 13]. Underlying an HMM is a basic Markov chain [14] In fact, an HMM is simply a Markov Source as defined by Gallager [15] a conditionally independent process on a Markov chain or, equivalently, a Markov chain viewed The authors are with the Information Systems Laboratory, Department of ....

R. Cole, L. Hirschman, L. Atlas, et al., "The challenge of spoken language systems: research directions for the nineties," IEEE Transactions on Speech and Audio Processing, Vol. 3, pp. 1-21, 1063-6676, Jan. 1995.

No context found.

Cole, R., et al. (1995). The Challenge of Spoken Language Systems: Research Directions for the Nineties. IEEE Transactions on Speech and Audio Processing, 3(1):1-21.

No context found.

Cole, R., et al. (1995). The Challenge of Spoken Language Systems: Research Directions for the Nineties. IEEE Transactions on Speech and Audio Processing, 3(1):1-21.

No context found.

Cole R., Hirschman L., Atlas L. et al., The Challenge of Spoken Language Systems: Research Directions for the Nineties. IEEE Trans. on Speech and Audio Processing, Vol. 3, No. 1, January 1995. pp. 1-21.

No context found.

R. Cole et al. "The challenge of spoken language systems: Research directions for Z. the nineties, IEEE Trans. Speech Audio Process., 3,1#21 1995 .

No context found.

Cole, R., et al. The Challenge of Spoken Language Systems: Research Directions for the Nineties. IEEE Transactions on Speech and Audio Processing, 3(1):1-21, January 1995.

No context found.

Cole, R., et al, "The Challenge of Spoken Language Systems: Research Directions for the Nineties", IEEE Trans. Speech and Audio Proc., 3, pp. 1-20, 1995.

No context found.

Cole R., et al, The challenge of spoken language systems: research directions of the 90's, IEEE Trans Speech and Audio Processing, 3, 1, 1995.

No context found.

Cole R., et al, The challenge of spoken language systems: research directions of the 90's, IEEE Trans Speech and Audio Processing, 3, 1, 1995.

No context found.

Cole, R. et al (1995) The Challenge of Spoken Language Systems: Research Directions for the Nineties, IEEE Transactions on Speech and Audio Processing, vol.3, No.1, January 1995, 1-21

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC