Effective Tamil Character Recognition Using Supervised Machine Learning Algorithms

Computational linguistics is the branch of in which techniques computer science are applied to analysis and synthesis language speech. The main goals computational include: Text-to- speech conversion, Speech-to-text conversion Translating from one another. A part Linguistics Character recognition. recognition has been active challenging research areas field image processing pattern methodology mainly focuses on recognizing characters irrespective difficulties that arises due variations writing style. aim this project perform character for complex structures south Indian ‘Tamil’ using a supervised algorithm increases accuracy novelty system it recognizes Predominant Tamil Language. proposed approach capable text where traditional systems fails, notably presence blur, low contrast, resolution, high noise, other distortions. This uses Convolutional Neural Network Algorithm able exact local features more accurately as they restrict receptive fields hidden layers be local. Networks great kind multi-layer neural networks back-propagation algorithm. used recognize visual patterns directly pixel images with minimal preprocessing. trained network classification. results show yields good rates.