Google Voice Search, a free service to recognize spoken web queries is currently available on most Android devices as well as on the iPhone/iPad as part of the Google mobile app. It has been steadily growing in traffic since its initial launch in the US in 2008 and since has been extended to general dictation of E-Mail, chat and other applications, now running in over 20 languages worldwide. In this talk I want to share some of the history of the development of the voice search technology, how we planned to develop it, what challenges we faced with different languages, how language models, acoustic models and dictionaries were developed, what maintenance is necessary to keep it working, and how we see Voice Search develop in the future. I will show some new features currently available only for a subset of languages and discuss how Google's speech technology can be used for any Android application.
Dr. Mike Schuster graduated in Electric Engineering from the Gerhard-Mercator University in Duisburg, Germany in 1993. After receiving a scholarship to study in Japan he spent a year in Japan to study Japanese in Kyoto and Fiber Optics in the Kikuchi laboratory at Tokyo University. His professional career in speech brought him to Advanced Telecommunications Research Laboratories in Kyoto, Nuance in the US and NTT in Japan where he worked on general machine learning and speech recognition research and development after getting his PhD at the Nara Institute of Science and Technology. Today he is part of the speech recognition group at Google where he focuses on machine learning applications and in particular on all aspects of making speech recognition technology available to the public, recently mostly through the VoiceSearch project.