AI Generated Cricket Score using NLP

To create an AI-based system for creating real- time cricket scorecards using live or recorded commentary, this study investigates the integration of Natural Language Processing (NLP), audio recognition, and machine learning approaches. Along with team names and venue information, users can stream live commentary using a microphone or upload audio files using the system’s user-friendly frontend, Streamlit. Speech recognition is used to process and turn the audio into text, which is subsequently tokenized and subjected to NLP techniques to extract important events like runs, wickets, and overs. The scorecard is updated continuously by appending this textual data to an already-existing match commentary file. Additionally, a T20 dataset is used to train a Random Forest-based machine learning model that uses the dynamically generated scorecard data to predict the final match score. By providing both live updates and predicted insights, the system seeks to improve user experience by delivering an automated, real-time cricket score creation tool. Index Terms—Natural language processing (NLP), speech recognition, artificial intelligence (AI), cricket scorecards, real- time commentary, machine learning, random forests, Text anal- ysis, tokenization, predictive modeling, Sports Data Extraction, Audio to Text Conversion, Automated Sports Analytics, and T20 Cricket Dataset In