Paper Details

Name: Bhawana, Ajay, Prince Sharma, Shivani Raj, Rishabh Aditya & Anirudh
Country: India

Year: 2026
Volume: Volume No: 13, January, Year: 2026 (Special Issue)
Page Number: 822-830
DOI: https://doi.org/10.5281/zenodo.19130622

ABSTRACT—
In the era of globalization, conferences and seminars across the world are now often delivered in many languages, with which it gets difficult to make smooth communication and sharing of information. This work describes a multilingual audio summarization system developed to increase convenience and ease for conference content. Through sophisticated natural language processing and machine learning technology, the system can transcribe, analyze, and create succinct summaries of multilingual audio recordings. It incorporates automatic speech recognition (ASR) for multilingual languages, then intelligent summarization algorithms to ensure real-time, contextually correct, and linguistically accurate summaries. This design enables crosslinguistic understanding so that participants can easily understand main insights irrespective of linguistic differences. Our findings establish the viability and utility of this system in enhancing information retrieval, reducing cognitive overload, and rendering key conference deliberations accessible to a global multilingual audience. Moreover, the presented framework extends its usage to many areas such as academia, business
meetings, and global collaborations. Connecting the gaps in languages, the solution empowers users with easy access to vital information, transforming the means through which global events enable communications and knowledge transfer.

Index Terms—Multi-language, Audio Summarization, Global Conferences, Natural Language Processing, Machine Learning, Speech Recognition, Real-time Summarization, Cross-Linguistic Communication, Information Access.