Our Urdu Text-to-Speech (TTS) system using transformer models to improve the quality and speed of speech synthesis. Traditional TTS systems often faced problems with slow responses and unnatural voice quality, making them difficult for Urdu speakers to use. Our solution focuses on speeding up text-to-speech conversion and enhancing voice naturalness by using our own collected dataset. With these improvements, our system delivers faster and more realistic speech, enabling smoother communication on various digital platforms
Keywords: TTS,NLP,DEEP LEARNING,URDU TTS,DATA SCIENCE
Tools: Python,Jupyter,Huggingface,OpenAI,API,Django,TensorFlow,PyTorch
Department: Department of Computer Science
Tools: Python,Jupyter,Huggingface,OpenAI,API,Django,TensorFlow,PyTorch
Department: Department of Computer Science
Project Poster
