Japan’s national phone company, NTT, has developed a voice-recognition system
that can analyse sentences as they are spoken. Conventional systems wait until
the end of a sentence before analysing it, which leads to delays. NTT says its
system can also tell from the inflection of the voice whether someone is making
a statement or asking a question. It plans to use the system for
ticket-reservation and information retrieval services.
More from New Scientist
Explore the latest news, articles and features
Popular articles
Trending New Scientist articles


