Language Weaver Expands Statistical Technology Capabilities; Hires New Exec
Tweet10/1/07 – Language Weaver, the Los Angeles-based developer of software for automatic language translation, says it has expanded its automated translation technology to include linguistics information and syntax rules in its statistical translation techniques. To help achieve this goal, the company has announced the hiring of David Savignac, Ph.D., as senior director of lexicography to head up the effort, the goal of which is to add less commonly taught languages to Language Weaver’s roster of automated translation modules. Savignac recently concluded a career of more than 30 years with the National Security Agency, where he had been founder and director of the Center for Applied Machine Translation.
Language Weaver says it has conducted several surveys to determine the language pairs that its commercial and government customers need for automated translation. In the government arena, these include many African, South Asian and Central Asian languages, and in commercial markets, Language Weaver says that companies want to be able to translate between not only English and other languages, but also German to Czech, Polish and Chinese, or French to Hindi and Chinese. Many of these languages have less parallel data available for training the automated translation engines, according to the company.
Language Weaver uses proprietary statistical translation algorithms that resulted from research and development at USC’s Information Sciences Institute. Earlier this year, it launched a consumer-focused subsidiary called Kontrib.
Founded in 2002, The company is backed by Palisades Ventures, the Tech Coast Angels and the Athenaeum Fund.
Language Weaver – www.languageweaver.com
