Text and speech recognition system using navigation information

• US 8,340,958 B2
• Filed: 01/25/2010
• Issued: 12/25/2012
• Est. Priority Date: 01/23/2009
• Status: Active Grant

• Current Assignee: Harman Becker Automotive Systems GmbH (Samsung Electronics Co. Ltd.)
• Original Assignee: Harman Becker Automotive Systems GmbH (Samsung Electronics Co. Ltd.)
• Inventors: Gruhn, Rainer, Riechert, Andreas Marcel
• Primary Examiner(s): Desir, Pierre-Louis
• Assistant Examiner(s): Sirjani, Fariba
• Application Number:
  US12/693,392
• Publication Number:
  US 20100191520A1
• Time in Patent Office: 1,065 Days
• Field of Search: 704 1- 9, 704/257, 704/270, 704/272, 704/275
• US Class Current: 704/9
• CPC Class Codes: G06F 3/0237 using prediction or retrieval techniques
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A system and method are provided for recognizing a user'"'"'s speech input. The method includes the steps for detecting the user'"'"'s speech input, recognizing the user'"'"'s speech input by comparing the speech input to a list of entries using language model statistics to determine the most likely entry matching the user'"'"'s speech input, and detecting navigation information of a trip to a predetermined destination, where the most likely entry is determined by modifying the language model statistics taking into account the navigation information. A system and method is further provided that takes into account navigation trip information to determine the most likely entry using language model statistics for recognizing text input.

Claims

1. A method for recognizing a user'"'"'s speech input comprising the following steps:

utilizing a navigation unit for;

detecting navigation information of a trip to a predetermined destination, and

modifying language model statistics in a language model database containing a list of entries for list entries that are in the language model database and that are contained in the navigation information;

detecting whether words in the navigation information are contained in the list of entries,

where any words in the navigation information that are not contained in the list of entries are added to the list of entries as new entries, and where the likelihood of the new entries is set higher than an average likelihood of the entries in the list of entries;

detecting the user'"'"'s speech input; and

recognizing the user'"'"'s speech input by comparing the user'"'"'s speech input to the list of entries using the modified language model statistics to determine the most likely entry best matching the user'"'"'s speech input,

where the navigation information contains at least one of the following information;

an overall travel time of the trip, the current travel time of the trip, and an expected arrival time.

2. The method of claim 1, where the language model statistics indicate a likelihood of the different entries in the list of entries, where the likelihood of a list entry contained in the navigation information is increased in the step of modifying language model statistics.

3. The method of claim 1, where the navigation information further contains at least one of the following information:

a starting location of the trip, a destination location of the trip, via points of the trip, and the current position.

4. The method of claim 1, further comprising the steps of:

detecting a current infotainment status;

determining list entries relating to said current infotainment status; and

increasing the likelihood of said infotainment-related list entries.

5. A method for supporting a text input in an input unit, the method comprising the steps of:

utilizing a navigation unit for;

detecting navigation information of a trip to a predetermined destination, and

modifying language model statistics in a language model database containing a list of entries for list entries that are in the language model database and that are contained in the navigation information;

detecting whether words in the navigation information are contained in the list of entries;

where any words in the navigation information that are not contained in the list of entries are added to the list of entries as new entries, and where the likelihood of the new entries is set higher than an average likelihood of the entries in the list of entries;

receiving text input at the input unit of an application that is not the navigation unit;

detecting at least a first part of the text input;

recognizing the text input by comparing said at least first part of the text input to a list of entries in the language model database based on the modified language model statistics in order to suggest a most likely candidate of a completed text input; and

suggesting said most likely candidate of the completed text input to a user of the input unit,

where the navigation information contains at least one of the following information;

an overall travel time of the trip, the current travel time of the trip, and an expected arrival time.

6. The method of claim 5, where the language model statistics indicate a likelihood of the different entries in the list of entries, where the likelihood of a list entry contained in the navigation information is increased in the step of modifying language model statistics.

7. The method of claim 5, where the navigation information further contains at least one of the following information:

a starting location of the trip, a destination location of the trip, via points of the trip, and the current position.

8. The method of claim 5, further comprising the steps of:

detecting a current infotainment status;

determining list entries relating to said current infotainment status; and

increasing the likelihood of said infotainment-related list entries.

9. The method of claim 5, further comprising the step of recognizing said first part of the text input using the navigation information based on the modified language model statistics when the first part of the text input is not unambiguous.

10. The method of claim 5, where the text input is input into the input unit in a hand-written form, where the text input is recognized taking into account the navigation information based on the modified language model statistics.

11. The method of claim 5, where the text is input in a phonetic representation and the complete text input is detected, where the phonetic representation is converted to different expressions having substantially the same phonetic representation, the most likely expression being selected taking into account the navigation information based on the modified language model statistics.

12. A speech recognition system recognizing a user'"'"'s speech input, comprising:

A navigation unit configured to generate navigation information of a trip to a predetermined destination;

a speech detector configured to detect the user'"'"'s speech input;

a language model database comprising a list of entries and language model statistics for the list of entries; and

a speech recognition unit configured to receive words in the navigation information from the navigation unit, and to adjust the language model statistics for entries in the list of entries that are contained in the navigation information, the speech recognition unit being configured to recognize the user'"'"'s speech input by comparing the speech input to the list of entries in the language model database using the modified language model statistics in order to determine the most likely entry best matching the user'"'"'s speech input,

where the speech recognition unit detects whether words in the navigation information are contained in the list of entries,

where any words in the navigation information that are not contained in the list of entries are added to the list of entries as new entries, and where the likelihood of the new entries is set higher than an average likelihood of the entries in the list of entries, and

where the navigation information contains at least one of the following information;

an overall travel time of the trip, the current travel time of the trip, and an expected arrival time.

13. A system for supporting a text input in an input unit, comprising:

a navigation unit configured to generate navigation information of a trip to a predetermined destination;

an input unit for inputting the text input;

a language model database comprising a list of entries and language model statistics for the list of entries; and

a text recognition unit configured to receive words in the navigation information from the navigation unit, and to adjust the language model statistics for entries in the list of entries that are contained in the navigation information, the text input recognition unit being configured to recognize the text input by comparing at least a first part of the text input to the list of entries in the language model database based on the modified language model statistics in order to suggest a most likely candidate of a completed text input,

where the text recognition unit detects whether words in the navigation information are contained in the list of entries,

where any words in the navigation information that are not contained in the list of entries are added to the list of entries as new entries, and where the likelihood of the new entries is set higher than an average likelihood of the entries in the list of entries, and

where the navigation information contains at least one of the following information;

an overall travel time of the trip, the current travel time of the trip, and an expected arrival time.