IWSLT was founded in 2004 as the first and premier international workshop and conference for spoken language translation. IWSLT is a forum for researchers from many disciplines to come together to discuss technology solutions for cross-lingual communication.

IWSLT emerged in 2004 from the international Consortium for Speech Translation Advanced Research (C-STAR) that was originally formed in 1992. C-STAR connected research labs around the world to develop speech translation systems, jointly.

C-STAR spent a first decade of collaboration to develop and connect speech translators for each country’s language, and to test their effectiveness in international demonstrators. Initial systems were domain and vocabulary limited and lacked robustness. They were thus not yet broadly deployable for every-day use, but the understandings gained from C-STAR, set the stage for more rigorous performance driven spoken language system development, that is now the core of IWSLT benchmarking activity.

Through systematic advances during the decades that followed the underlying technologies, Speech Recognition, Spoken Language Translation, Interfaces and integration in different use-cases (e.g. consecutive and simultaneous interpretation) have now advanced to performance levels that drive broad deployment of the technology and now make cross-lingual communication possible, worldwide

Past Editions