Optimization of stocking density for environmental-friendly improved extensive shrimp farming system in south-west part of Bangladesh

An experimental culture practice of P. monodon on extension approach was conducted in two brackish water earthen ponds of Demonstration Farm and Training Center (DFTC), Kaliganj, Satkhira. The experiment was aimed to provide farmers with appropriate technology that can immediately improve pond yield with keeping the environment in friendly condition. For optimization of stocking density of a cost effective environmental friendly improved extensive shrimp farming, the ponds were stocked with coastal river post larvae of P. monodon at the stocking rates of 2 pls/m² and 2.5 pls/m² without supplementary feeding. To control experimental error another five farmer's gher were used as replicates of each demo-pond. Considering the farmers buying ability, cost of inputs and other facilities kept minimal. The impact of stocking density was evaluated on the basis of growth, survival rate, production and economic return. Better production (average 299.01 kg/ha) with same survival rate (39.33%) were found with a stocking density of 2.5 pls/m² without causing any deterioration in the culture environment.

Continuous speech recognition for the TIMIT database using neural networks

Four types of neural networks which have previously been established for speech recognition and tested on a small, seven-speaker, 100-sentence database are applied to the TIMIT database. The networks are a recurrent network phoneme recognizer, a modified Kanerva model morph recognizer, a compositional representation phoneme-to-word recognizer, and a modified Kanerva model morph-to-word recognizer. The major result is for the recurrent net, giving a phoneme recognition accuracy of 57% from the si and sx sentences. The Kanerva morph recognizer achieves 66.2% accuracy for a small subset of the sa and sx sentences. The results for the word recognizers are incomplete.

MMI training for continuous phoneme recognition on the TIMIT database

This paper reports our experiences with a phoneme recognition system for the TIMIT database which uses multiple mixture continuous density monophone HMMs trained using MMI. A comprehensive set of results are presented comparing the ML and MMI training criteria for both diagonal and full covariance models. These results using simple monophone HMMs show clear performance gains achieved by MMI training, and are comparable to the best reported by others including those which use context-dependent models. In addition, the paper discusses a number of performance and implementation issues which are crucial to successful MMI training.

Optimization of semiconductor optical amplifiers for all optical wavelength conversion

In recent years a variety of experimental and theoretical work has been reported on the use of semiconductor optical amplifiers for high speed wavelength conversion. However little work has addressed the dynamic limitations of this conversion process in detail with a view to device optimization. In this paper, a detailed study of the conversion process is carried out in order to optimize device parameters and drive conditions for increased conversion speed and improved modulation index.