Articles written in Sadhana
Volume 19 Issue 1 February 1994 pp 1-3 Artificial Intelligence And Expert Systems
Volume 19 Issue 1 February 1994 pp 147-169 Artificial Intelligence And Expert Systems
This paper discusses the significance of segmental and prosodic knowledge sources for developing a text-to-speech system for Indian languages. Acoustic parameters such as linear prediction coefficients, formants, pitch and gain are prestored for the basic speech sound units corresponding to the orthographic characters of Hindi. The parameters are concatenated based on the input text. These parameters are modified by stored knowledge sources corresponding to coarticulation, duration and intonation. The coarticulation rules specify the pattern of joining the basic units. The duration rules modify the inherent duration of the basic units based on the linguistic context in which the units occur. The intonation rules specify the overall pitch contour for the utterance (declination or rising contour), fall-rise patterns, resetting phenomena and inherent fundamental frequency of vowels. Appropriate pauses between syntactic units are specified to enhance intelligibility and naturalness.
Volume 19 Issue 2 April 1994 pp 189-238
This tutorial article deals with the basics of artificial neural networks (ANN) and their applications in pattern recognition. ANN can be viewed as computing models inspired by the structure and function of the biological neural network. These models are expected to deal with problem solving in a manner different from conventional computing. A distinction is made between pattern and data to emphasize the need for developing pattern processing systems to address pattern recognition tasks. After introducing the basic principles of ANN, some fundamental networks are examined in detail for their ability to solve simple pattern recognition tasks. These fundamental networks together with the principles of ANN will lead to the development of architectures for complex pattern recognition tasks. A few popular architectures are described to illustrate the need to develop an architecture specific to a given pattern recognition problem. Finally several issues that still need to be addressed to solve practical problems using ANN approach are discussed.
Volume 21 Issue 1 February 1996 pp 5-20 Recent Results In Signal Processing And Communication
The objective of this paper is to demonstrate the importance of position of the analysis time window in time-frequency analysis of speech signals. Speech signals contain information about the time varying characteristics of the excitation source and the vocal tract system. Resolution in both the temporal and spectral domains is essential for extracting the source and system characteristics from speech signals. It is not only the resolution, as determined by the analysis window in the time domain, but also the position of the window with respect to the production characteristics that is important for accurate analysis of speech signals. In this context, we propose an event-based approach for speech signals. We define the occurrence of events at the instants corresponding to significant excitation of the vocal tract system. Knowledge of these instants enable us to place the analysis window suitably for extracting the characteristics of the excitation source and the vocal tract system even from short segments of data. We present a method of extracting the instants of significant excitation from speech signals. We show that with the knowledge of these instants it is possible to perform prosodic manipulation of speech and also an accurate analysis of speech for extracting the source and system characteristics.
Volume 21 Issue 3 June 1996 pp 395-413 Intelligent systems
The objective of this study is to explore the possibility of capturing the reasoning process used in bidding a hand in a bridge game by an artificial neural network. We show that a multilayer feedforward neural network can be trained to learn to make an opening bid with a new hand. The game of bridge, like many other games used in artificial intelligence, can easily be represented in a machine. But, unlike most games used in artificial intelligence, bridge uses subtle reasoning over and above the agreed conventional system, to make a bid from the pattern of a given hand. Although it is difficult for a player to spell out the precise reasoning process he uses, we find that a neural network can indeed capture it. We demonstrate the results for the case of one-level opening bids, and discuss the need for a hierarchical architecture to deal with bids at all levels.