Voice Recognition for Children and Toddlers Voices using Wav2vec2

This project, Kidzinya, develops an assistive AI system for real-time voice recognition and speech-to-text tailored specifically to children and toddlers. It addresses unique challenges such as high pitch and speech variability, using fine-tuned Wav2Vec2 models to achieve up to 96% accuracy. The system provides immediate visual feedback to support early language development. Key features include speech-to-text interaction, handwriting recognition, and an adaptive learning path. Built with Python, TensorFlow, Flutter, and Hugging Face datasets, Kidzinya outperforms similar apps like Khan Academy Kids and ABCmouse in child-specific voice recognition capabilities. This work was conducted at Arab International University (AIU), Syria. The official website of the university is: https://www.aiu.edu.sy