System, method and computer program product for point-to-point voice-enabled driving directions

• US 6,789,065 B2
• Filed: 01/24/2001
• Issued: 09/07/2004
• Est. Priority Date: 01/24/2001
• Status: Expired due to Term

• Current Assignee: Cerence Inc., Cerence Operating Company (Cerence Inc.)
• Original Assignee: Bevocal Incorporated (Microsoft Corporation)
• Inventors: Stone, Kevin, Berner, Mikael, Guerra, Lisa, Gaitonde, Kavita
• Primary Examiner(s): ABEBE, DANIEL DEMELASH
• Application Number:
  US09/769,903
• Publication Number:
  US 20020133351A1
• Time in Patent Office: 1,322 Days
• Field of Search: 704/246, 704/251, 704/270, 704/235, 704/260, 704/272, 704/275, 701/200, 701/201, 701/202, 701/209, 701/211, 701/213
• US Class Current: 704/275
• CPC Class Codes: G01C 21/3608 using speech input, e.g. using speech recognition
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A system, method and computer program product are afforded for providing voice-enabled driving directions. Initially, an utterance representative of a destination address is received. Thereafter, the utterance is transcribed utilizing a speech recognition process. An origin address is then determined. A database is subsequently for queried generating driving directions based on the destination address and the origin address.

Claims

1. A method for providing voice-enabled driving directions, comprising:

(a) receiving an utterance representative of a destination address including a street over a network;

(b) transcribing the utterance utilizing a speech recognition process, wherein the speech recognition process includes;

(i) querying one of a plurality of databases of different types selected from the group consisting of static, dynamic, web server, and file system,

(ii) dynamically retrieving grammars utilizing protocols based on the type of the database,

(iii) determining whether the grammars are retrieved from a first one of the databases during a first attempt, and

(iv) retrieving the grammars from a second one of the databases upon the failure of the first attempt,

(v) wherein at least one of the databases is established based on a first source with a plurality of city names and associated zip codes, and a second source with a plurality of street names, the establishing being validated using a third source,

(vi) wherein at least one of the databases assigns a probability to each of the grammars which indicates a prevalency of use of the grammars, the probability being determined using statistical data corresponding to use of streets associated with the grammars, and the utterance being matched with one of the grammars in the database based at least in part on the probability;

(c) determining an origin address, wherein the origin address is determined utilizing the speech recognition process; and

(d) querying a database for generating driving directions based on the destination address and the origin address.