CTS-Synthesizer: Handwritten Character to Speech Conversion using CNN and Google Text-to-Speech Synthesizer

In the ever-evolving field of artificial intelligence and computer vision, a Handwritten Character-to-speech (CTS) Synthesizer has emerged as an integral component within human-like interactions. Although research has been studied to recognize either digit or character recognition, this paper presents a machine learning-based method for real-time converting handwritten alphanumeric (refer to both English characters and digits) words to speech. The approach utilizes a convolutional neural network (CNN) for precise handwritten character recognition and integrates the Google Text-to-Speech (TTS) synthesizer for converting recognized characters into speech. The proposed model is trained and tested using a combination of two well-known handwritten datasets (one for characters and one for digits). The CNN architecture is employed to effectively extract features and classify characters and digits from pre-processed images. Once the text is recognized, it is then processed by Google Text-to-Speech to generate natural-sounding speech. The results demonstrate that the proposed model achieves high accuracy in predicting 36 English handwritten characters and digits, with an overall accuracy of 99.22%. The proposed system can be applied in various domains and made a valuable contribution to the field of Natural Language Processing (NLP), offering significant potential applications in assistive assistive technology for visually impaired individuals and automated data entry systems and beyond.