Google Receives Patent For Voice Search

Google was granted a patent (# 7,027,987) today that covers converting voice to search queries. While there is not currently a mobile version of the technology, it seems like the most obvious extension for the technology. Google’s technology converts natural spoken language into a boolean search query. From the patent’s abstract:

The system receives a voice search query from a user, derives one or more recognition hypotheses, each being associated with a weight, from the voice search query, and constructs a weighted boolean query using the recognition hypotheses. The system then provides the weighted boolean query to a search system and provides the results of the search system to a user.

There are several players in voice-to-text (VoiceSignal) and voice search (V-Enable, Promtu) in the mobile space. Google is the only major search engine to publicly test a voice interface.

1 Response to “Google Receives Patent For Voice Search”


  1. 1 Sam

    Nothing really new here. Google had been working on voice-enabled search since 2001. It already filed a patent on voice search in February 2001 (http://patft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=7027987.PN.&OS=PN/7027987&RS=PN/7027987) and had a beta voice search interface from 2003 to 2005 within Google Labs (http://labs1.google.com/gvs.html).

    The earlier demo was taking search queries and returned results only over the phone. Most promising real-world applications would have results sent to a mobile phone screen or in-car system, especially for local search. In a 2004 interview, Google said it envisioned a voice interface to aid in “everything from driving directions to finding groceries in a supermarket”.

    The latest patent does go one step further though. Beside mentioning “receiving a voice search query from a user” like in the 2001 patent, it also mentions “providing the weighted boolean query to a search system”, probably to deal with high word error rates and limited sets of word choices used in some voice interfaces.

    In short, Google is still building up its infrastructure for voice recognition applied to search. Their focus is now to reduce the error rate. When they will be satisfied with the result, expect to see voice interfaces leveraged across multiple products. Barry Schnitt from Google commented on the latest patent: “Some of those ideas later mature into real products or services, some don’t. Prospective product announcements should not be inferred from our patent applications.”

Leave a Reply