Transcribing The TV News Archive's 2.5-Million-Hour Quarter-Century Archive Of Global Television News

The Internet Archive's Television News Archive today preserves global television news coverage encompassing selections from 108 channels spanning 50 countries and territories in at least 35 languages and dialects over 23 years on 5 continents. We are immensely excited to announce today that in collaboration with the Archive, we have machine transcribed its complete quarter-century global archive totaling 149.7 million minutes (2.5 million hours) of uncaptioned airtime using GCP's state-of-the-art Universal Speech Model Chirp ASR model in what we believe is one of the largest scholarly applications of multilingual speech recognition to global television news. Over the coming months we'll be working with the Archive to make the complete 2.5-million-hour quarter-century 50-country global archive keyword searchable on the Archive's website and in the Television Explorer for advanced analytics using these machine transcripts, allowing scholars for the first time to look across the world's coverage of events from local to global. Moreover, Chirp's native support for code switching and truly multilingual broadcasts means that for the first time scholars will be able to search across all of the languages represented in the Archive and explore the underlying patterns of spoken multilingualism and the role of language in storytelling across the world. This immense milestone caps more than a decade of our collaborations of the Archive and its own work with scholars and journalists from across the world that have already collectively yielded more than 2,000 articles and research papers. As the TV News Archive officially turns 12 this year and for the first time since it captured global reaction to 9/11 almost 23 years ago, scholars will for the first time be able to search this immense global archive to understand how our planet's biggest stories of the past quarter-century have been covered across the world.

Stay tuned for a wealth of blog posts over the coming months as we dive into this massive global archive to explore a quarter century of global television news and slowly make the complete Archive searchable for the first time since its creation!