A new patent has provided a glimpse of Google's pending voice-activated search engine.
The patent involves "a voice interface for search engines. Through the use of a language model, phonetic dictionary and acoustic models, a server generates an n-best hypothesis list or word graph."
Voice-in, text-out searching has been around as a concept at least since IBM promoted it as the way to go beyond the limitations of ever smaller cell phones in 1999. Google is not the only one to be working in the area - News.com reports a voice-activated Internet system works with speech-to-text conversion software called Maestro was unveiled in Israel in May of last year.
Maestro is said to convert spoken search requests into a list of query-friendly ASCII words which are relayed to a search engine, with results returned to the searcher audibly.
A Google spokesman told News.com that "prospective product announcements should not be inferred from our patent applications". The Google Voice Search demo is currently not available.