A-CLAPS: Automatic Correction of Language and Pronunciation Errors in Slide-based Presentations

This study introduces A-CLAPS, an AI-driven system designed to seamlessly enhance the quality of slide-based video presentations. A-CLAPS leverages the latest advances in Automatic Speech Recognition (ASR), Text-to-Speech (TTS), voice cloning, and Large Language Models (LLMs) to correct grammatical errors, address mispronunciations, and adjust speech tempo, all while preserving the speaker’s original voice. The system’s effectiveness is assessed through quantitative and qualitative comparisons of the original and AI-enhanced videos, using metrics such as speech rate, word and sentence counts, and audio fidelity measures like MFCC and Spectral Convergence. Experimental results show that A-CLAPS improves linguistic accuracy and presentation quality while preserving the speaker’s unique vocal characteristics and style. By making professional presentations more accessible and comprehensible, A-CLAPS has a broad impact on advancing fair and inclusive speech science and technology.