If it seems on first glace that such a patent is simply taking an existing idea -- speech recognition -- and applying it to another existing technology -- search -- Google has a rejoinder to that. According to its patent application:
"Current speech recognition technology has high word error rates for large vocabulary sizes. … Current voice interfaces to search engines address the problems by limiting the scope of the voice queries to a very narrow range. At every turn, the user is prompted to select from a small number of choices. For example, at the initial menu, the user might be able to choose from "news," "stocks," "weather," or "sports." After the user chooses one category, the system offers another small set of choices. By limiting the number of possible utterances at every turn, the difficulty of the speech recognition task is reduced to a level where high accuracy can be achieved.
This approach results in an interactive voice system that has a number of severe deficiencies. It is slow to use, since the user must navigate through may levels of voice menus. If the user's information need does not match a predefined category, then it becomes very difficult or impossible to find the information desired. Moreover, it is often frustrating to use, since the user must adapt his/her interactions to the rigid, mechanical structure of the system.
Therefore, there exists a need for a voice interface that is effective for search engines."
Interestingly, Google appears to have already conducted some small-scale testing on this approach, in the form of its Google Voice Search Demo. The technique promises to let users "search on Google by voice with a simple telephone call." Unfortunately, the test demo is currently unavailable. The status of the project is unclear; Google suggests users check back at some unspecified future time. There are some message threads on the technology, but all are low activity, suggesting not much is going on right now.
Another effort of note is apparently ongoing at Microsoft Research, where a project dubbed Speech Technology (Asia) is looking into voice-driven search for Chinese-speaking users. This is a particularly apt area for speech recognition, given the pictographic complexity of the Chinese character set.
Way off the beaten path, a search engine called Midomi purports to let you search for music files by humming or singing part of the song into your computer. (I'll try it after I close the door to my office.)
Obviously, efforts such as those of Google and Microsoft are flying below the radar screen (after all, the patent I'm so excited about above was awarded nearly a year ago. But the appeal of voice-driven search is intuitively obvious and merits, if nothing else, a big shout-out, at least from this blogger.