Foreign Language Automated Information Retrieval (FLAIR)/Machine Translation For English Retrieval of Information In Any Language (MATERIAL)
Abstract
This is the final report for the FLAIR team under the IARPA MATERIAL program. The objective of the program was to develop the technologies for cross-lingual information retrieval from speech or text documents in foreign languages. The retrieved relevant documents, based on an English query, are then summarized in English for presentation to the analyst. Progress was made in the four requisite technologies: Automatic Speech Recognition (ASR), Machine Translation (MT), Cross-Lingual Information Retrieval (CLIR), and Summarization. The report summarizes the advances made in the four technologies throughout the program. The project resulted in a number of deliveries of software and data to the Government. The MT and CLIR technologies have been transitioned to the Raytheon BBN Multi-Media Monitoring System (M3S) for eventual deployment in Government installations.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 2021
- Accession Number
- AD1165370
Entities
People
- Damianos Karakos
- David Akodes
- John Makhoul
- Le Zhang
- Lee Tarlin
- Manaj Srivastava
- Rich Schwartz
- Sanjay K. Gouda
- William Hartmann
Organizations
- RTX