SAIL LABS Technology, one of the world’s leading innovators in the field of speech technology has been selected to present “CAVA” (Continuous Automatic Vocabulary Adaptation) at SpeechTEK 2008, the world’s biggest speech technology conference and exhibition, at the New York Marriott Marquis.
Mark Pfeiffer, Vice President Business Development & Communications at SAIL LABS Technology, has been selected to speak: “We are honored to have been chosen to present our latest ground-breaking development – CAVA (Continuous Automatic Vocabulary and language model Adaptation). CAVA is a framework that automatically adapts and changes automatic speech recognition (ASR) language models over time to current affairs. The motivation behind developing a framework like this was to be able to respond to dynamic environments: Speech recognition systems age over time as ASR remains the same and “currently used” vocabularies change; i.e. people, places and words are introduced into the current vocabulary or fade away over time,” said Mr. Pfeiffer. “A current example is the South Ossetia conflict: most language models would not have had the places or acting people of this war included in the vocabulary. With CAVA – within a short period – these would have been added automatically without manual intervention by the user. CAVA is the result of us listening to the needs of our clients, who heavily rely on having an easy to operate system that gives them access to current Open Source Information (OSINF) from around the world in many languages. With CAVA and the other SAIL LABS Systems, they can now focus even more on their mission, which is gathering and analysing OSINF and turning it into actionable intelligence.”
“We have developed CAVA with the vision of creating a technology to allow for long-term, unsupervised adaptation of a system deployed in the real-world. SpeechTEK is the perfect venue to share this with our industry and our technology expertise with other high-profile experts and a wide range of organizations,” concluded Mark Pfeiffer.
“CAVA” is scheduled in TRACK C: ADVANCED SPEECH TECHNOLOGIES SYMPOSIUM (session C201-Natural Language Processing) on Tuesday, August 19.
Share on Social Media