Newsbridge Reveals New Language Enhancements for MXT-1 Generative AI Indexing Technology


Newsbridge announced new language enhancements for MXT-1, the company’s multimodal and generative AI indexing technology.

Newsbridge says its core AI indexing tech uses natural language models to generate human-like descriptions of video content, with the power to index more than 500 hours of video per minute.

Newsbridge claims the MXT-1 is now able to describe what’s happening in every shot of a video, as well as summarize the video in its entirety. These scene descriptions and media summaries are available in English, French, Spanish, German, Portuguese and Arabic, with more languages to come.

This enables users to choose the language they are most comfortable with, improving their engagement and understanding of the video content, according to Newsbridge. In addition, MXT-1 enables multilingual transcription. Media organizations can automatically detect and transcribe a single media asset when more than one language is spoken.

Newsbridge says the new multi-language capabilities complement MXT-1’s “already powerful” transcription and neural translation functionalities, which are available in more than 130 languages. Through MXT-1, media companies can translate video transcriptions into hundreds of languages in near real-time to extend the global reach of their video content.

Newsbridge says MXT-1 generative AI indexing technology is used by leading media organizations, including “Le Parisien,” a French newspaper producing both local and global news.