Navigation Device and Address Input Method Thereof

• US 20070244629A1
• Filed: 03/20/2007
• Published: 10/18/2007
• Est. Priority Date: 04/17/2006
• Status: Active Grant

• Current Assignee: Clarion Co., Ltd. (Faurecia SA)
• Original Assignee: Xanavi Informatics Corporation (Hitachi, Ltd.)
• Inventors: Hirayama, Yoshikazu
• Granted Patent: US 7,295,923 B2
• Time in Patent Office: Days
• US Class Current: 701/200
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

Voice input of an address, which is user-friendly and has improved recognition result accuracy. First, a whole street name (full name) is uttered and voice recognition is tried. Only if the voice recognition fails, at least one letter of a character string expressing the street name is uttered, and the letter is used to narrow down candidates for the street name.

Claims

1. A navigation device, comprising:

a storage means, which stores street names;

a first street name receiving means, which receives voice input of a whole street name;

a first street name specifying means, which specifies a street name that best matches the voice received by the first street name receiving means among the street names stored in the storage means;

a confirmation means, which receives confirmation of the street name specified by the first street name specifying means;

a letter receiving means, which receives voice input of at least one letter from a character string expressing the street name specified by the first street name specifying means, when the confirmation means has not received confirmation of the specified street name;

a letter specifying means, which recognizes the voice input received by the letter receiving means, so as to specify a letter;

an extraction means, which extracts candidate street names from the street names stored in the storage means, by using the letter specified by the letter specifying means;

a second street name receiving means, which receives voice input of the whole street name after the letter receiving means receives the voice input;

a second street name specifying means, which specifies a street name that best matches the voice input received by the second street name receiving means, from among the street names extracted by the extraction means; and

an output means, which outputs the street name specified by the second street name specifying means.

5. A navigation device of claim 1, wherein:

the letter specifying means obtains reliability for each alphabetical letter from the voice input received by the letter receiving means and extracts a letter having reliability higher than or equal to a predetermined value.

2. A navigation device, comprising:

a storage means, which stores street names;

a first street name receiving means, which receives voice input of a whole street name;

a first street name specifying means, which specifies a street name that best matches the voice input received by the first street name receiving means, from among the street names stored in the storage means;

a confirmation means, which receives confirmation of the street name specified by the first street name specifying means;

a letter receiving means, which receives voice input of a first letter of a street name when the confirmation means has not received confirmation of the specified street name;

a letter specifying means, which recognizes the voice input received by the letter receiving means, so as to specify a letter;

an extraction means, which extracts street names beginning with the letter specified by the letter specifying means, from the street names stored in the storage means;

a second street name receiving means, which receives voice input of the whole street name after the letter receiving means receives the voice input;

a second street name specifying means, which specifies a street name that best matches the voice input received by the second street name receiving means from among the street names extracted by the extraction means; and

an output means, which outputs the street name specified by the second street name specifying means.

3. A navigation device, comprising:

a storage means, which stores street names by state name and city name;

a state name receiving means, which receives voice input of a state name;

a state name specifying means, which recognizes the voice input received by the state name receiving means, so as to specify a state name;

a city name receiving means, which receives voice input of a city name;

a city name specifying means, which specifies a city name that belongs to the state name specified by the state name specifying means and best matches the voice input received by the city name receiving means;

a first street name receiving means, which receives voice input of a street name;

a first street name specifying means, which specifies a street name that belongs to the city name specified by the city name specifying means and best matches the voice input received by the first street name receiving means among the street names stored in the storage means;

a confirmation receiving means, which receives confirmation of the street name specified by the first street name specifying means;

a letter receiving means, which receives voice input of a first letter of a street name when the confirmation receiving means has not received confirmation of the specified street name;

a letter specifying means, which recognizes the voice input received by the letter receiving means, so as to specify a letter;

a street name extraction means, which extracts street names that belong to the city name specified by the city name specifying means and begin with the letter specified by the letter specifying means, among the street names stored in the storage means;

a second street name receiving means, which receives voice input of the whole street name after the letter receiving means receives the voice input;

a second street name specifying means, which specifies a street name that best matches the voice input received by the second street name receiving means, from among the street names extracted by the street name extraction means; and

an output means, which outputs the street name specified by the second street name specifying means.

4. A navigation device, comprising:

a storage means, which stores street names;

a first street name receiving means, which receives voice input of a whole street name;

a first street name specifying means, which specifies a street name that best matches the voice input received by the first street name receiving means, from among the street names stored in the storage means;

a confirmation means, which receives confirmation of the street name specified by the first street name specifying means;

a letter receiving means, which receives, by voice input, a correct letter to replace a wrong letter in a character string expressing the street name specified by the first street name specifying means, together with a position of the wrong letter in the character string, when the confirmation means has not received confirmation of the specified street name;

a letter specifying means, which recognizes the voice input received by the letter receiving means, so as to specify the correct letter;

a character string acquisition means, which acquires, from a character string that expresses the street name specified by the first street name specifying means, a character string extending as far as the correct letter in a character string resulting from replacing the wrong letter by the correct letter;

an extraction means, which extracts street names beginning with the acquired character string, from among the street names stored in the storage means;

a second street name receiving means, which receives voice input of the whole street name after the letter receiving means receives the voice input;

a second street name specifying means, which specifies a street name that best matches the voice input received by the second street name receiving means from among the street names extracted by the extraction means; and

an output means, which outputs the street name specified by the second street name specifying means.

6. An address input method for a navigation device, wherein the navigation device has a storage means which stores street names, the address input method comprising:

a first street name receiving step, in which voice input of a whole street name is received;

a first street name specifying step, in which a street name that best matches the voice input received in the first street name receiving step is specified, from among the street names stored in the storage means;

a confirmation step, in which confirmation of the street name specified in the first street name specifying step is received;

a letter receiving step, in which voice input of at least one letter of a character string expressing the street name specified by the first street name specifying step is received, when confirmation of the specified street name has not been received in the confirmation step;

a letter specifying step, in which the voice input received in the letter receiving step is recognized so as to specify a letter;

an extraction step, in which candidate street names are extracted from the street names stored in the storage means, by using the letter specified in the letter specifying step;

a second street name receiving step, in which voice input of the whole street name is received after the voice input is received in the letter receiving step;

a second street name specifying step, in which a street name that best matches the voice input received in the second street name receiving step, from among the street names extracted in the extraction step; and

an output step, in which the street name specified in the second street name specifying step is outputted.