Speech Recognition Device, Speech Recognition Method, and Program

• US 20080167872A1
• Filed: 06/02/2005
• Published: 07/10/2008
• Est. Priority Date: 06/10/2004
• Status: Active Grant

• Current Assignee: Sovereign Peak Ventures, LLC (Dominion Harbor Enterprises, LLC)
• Original Assignee: Panasonic Corporation (Panasonic Holdings Corporation)
• Inventors: Inoue, Tsuyoshi, Tsuzuki, Takashi, Okimoto, Yoshiyuki
• Granted Patent: US 7,813,928 B2
• Time in Patent Office: Days
• US Class Current: 704/251
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/0631 Creating reference templates; Clustering
• CPC Class Codes: G10L 2015/221 Announcement of recognition results

Abstract

A speech recognition device that is capable of presenting, to a user in an easy-to-understand manner, whether or not the user'"'"'s utterance is a word unregistered in a speech recognition dictionary and whether or not the utterance should be repeated due to a recognition error includes: a speech recognition vocabulary storage unit (102) which defines vocabulary for speech recognition; a speech recognition unit (101) which checks the uttered speech against the registered words; a reference similarity calculation unit (103) which calculates a similarity between the uttered speech and a combination of acoustic units, which are subwords; an unregistered word judgment unit (104) which judges, based on the result of the check by the speech recognition unit (101) and a result of the calculation performed by the reference similarity calculation unit (103), whether the uttered speech is a registered word or an unregistered word; an unregistered word storage (106) which stores unregistered words; an unregistered word candidate search unit (105) which searches the unregistered word storage unit (106) for unregistered word candidates that are likely to correspond to the uttered speech, in the case where the unregistered word judgment unit (104) judges the uttered speech to be an unregistered word; and a result display unit (107) which displays the result of the search.

Claims

1. A speech recognition device that recognizes uttered speech, comprising:

a speech recognition word storage unit operable to define vocabulary for speech recognition and to store the vocabulary as registered words;

a speech recognition unit operable to check the uttered speech against the registered words stored in said speech recognition word storage unit;

a reference similarity calculation unit operable to calculate a similarity between the uttered speech and a combination of acoustic units, which are subwords;

an unregistered word judgment unit operable to judge, based on a result of the check by said speech recognition unit and a result of the calculation performed by said reference similarity calculation unit, whether the uttered speech is a registered word which is stored in said speech recognition word storage unit or an unregistered word which is not stored in said speech recognition word storage unit;

an unregistered word storage unit operable to store unregistered words;

an unregistered word candidate search unit operable to search for an unregistered word candidate that is likely to correspond to the uttered speech, based on the uttered speech, among the unregistered words stored in said unregistered word storage unit, in the case where said unregistered word judgment unit judges the uttered speech to be an unregistered word; and

a result display unit operable to display a result of the search.

2. The speech recognition device according to claim 1,

wherein said unregistered word candidate search unit is operable to search for plural unregistered word candidates among the unregistered words stored in said unregistered word storage unit.

3. The speech recognition device according to claim 1,

wherein said unregistered word storage unit is operable to store the unregistered words according to categories into which the unregistered words are classified.

4. The speech recognition device according to claim 3, further comprising

an unregistered word class judgment unit operable to judge a category of the unregistered word based on the uttered speech.

5. The speech recognition device according to claim 4,

wherein said unregistered word candidate search unit is operable to search for the unregistered word candidates among the categories into which the unregistered words are classified in said unregistered word storage unit, based on a result of the judgment by said unregistered word class judgment unit.

6. The speech recognition device according to claim 3, further comprising

an information acquisition unit operable to acquire information regarding the categories,

wherein said unregistered word candidate search unit is operable to search for the unregistered word candidates among the categories into which the unregistered words are classified in said unregistered word storage unit, based on the information acquired by said information acquisition unit.

7. The speech recognition device according to claim 1,

wherein said result display unit is operable to display the result of the search by said unregistered word candidate search unit, by excluding the registered words stored in said speech recognition word storage unit from the result of the search.

8. The speech recognition device according to claim 1,

wherein said unregistered word candidate search unit is operable to search for the unregistered word candidate by calculating an unregistered word score indicating a degree of similarity to the uttered speech.

9. The speech recognition device according to claim 8,

wherein said result display unit is operable to display the unregistered word candidate and the calculated unregistered word score, as the result of the search.

10. The speech recognition device according to claim 9,

wherein said result display unit is operable to change the display of the unregistered word candidate according to the calculated unregistered word score.

11. The speech recognition device according to claim 1,

wherein the unregistered words stored in said unregistered word storage unit are updated under a predetermined condition.

12. The speech recognition device according to claim 11, further comprising

a communication unit operable to communicate with an unregistered word server that stores unregistered words which are not stored in said unregistered word storage unit,

wherein the unregistered words stored in said unregistered word storage unit are updated when said communication unit receives the unregistered words from the unregistered word server.

13. The speech recognition device according to claim 1,

wherein the registered words stored in said speech recognition word storage unit are updated under a predetermined condition.

19. The speech recognition device according to claim 2,

wherein said unregistered word storage unit is operable to store the unregistered words according to categories into which the unregistered words are classified.

14. A speech recognition system for recognizing uttered speech, comprising:

a speech recognition device that recognizes uttered speech; and

an unregistered word search server that searches for an unregistered word which is not stored in said speech recognition device,

wherein said speech recognition device includes;

a speech recognition word storage unit operable to define vocabulary for speech recognition and to store the vocabulary as registered words;

a speech recognition unit operable to check the uttered speech against the registered words stored in said speech recognition word storage unit;

a reference similarity calculation unit operable to calculate a similarity between the uttered speech and a combination of acoustic units, which are subwords;

an unregistered word judgment unit operable to judge, based on a result of the check by said speech recognition unit and a result of the calculation performed by said reference similarity calculation unit, whether the uttered speech is a registered word which is stored in said speech recognition word storage unit or an unregistered word which is not stored in said speech recognition word storage unit;

a search request sending unit operable to send, to said unregistered word search server, a search request to search for an unregistered word candidate that is likely to correspond to the uttered speech, in the case where said unregistered word judgment unit judges the uttered speech to be an unregistered word;

a search result receiving unit operable to acquire a result of the search for the unregistered word candidate from said unregistered word search server; and

a result display unit operable to display a result of the search, and

wherein said unregistered word search server includes;

an unregistered word storage unit operable to store unregistered words;

a search request receiving unit operable to receive the search request from said search request sending unit;

an unregistered word candidate search unit operable to search for an unregistered word candidate that is likely to correspond to the uttered speech, based on the uttered speech, among the unregistered words stored in said unregistered word storage unit, in the case where said search request receiving unit receives the search request; and

a search result sending unit operable to send the result of the search to said speech recognition device.

15. A speech recognition device that recognizes uttered speech, in a speech recognition system including said speech recognition device and an unregistered word search server that searches for an unregistered word which is not stored in said speech recognition device, said speech recognition device comprising:

a speech recognition word storage unit operable to define vocabulary for speech recognition and to store the vocabulary as registered words;

a speech recognition unit operable to check the uttered speech against the registered words stored in said speech recognition word storage unit;

a reference similarity calculation unit operable to calculate a similarity between the uttered speech and a combination of acoustic units which are subwords

an unregistered word judgment unit operable to judge, based on a result of the check by said speech recognition unit and a result of the calculation performed by said reference similarity calculation unit, whether the uttered speech is a registered word which is stored in said speech recognition word storage unit or an unregistered word which is not stored in said speech recognition word storage unit;

a search request sending unit operable to send, to the unregistered word search server, a search request to search for an unregistered word candidate that is likely to correspond to the uttered speech, in the case where said unregistered word judgment unit judges the uttered speech to be an unregistered word;

a search result receiving unit operable to acquire a result of the search for the unregistered word candidate from the unregistered word search server; and

a result display unit operable to display the result of the search.

16. (canceled)

17. A program used for a speech recognition device that recognizes uttered speech said program causing a computer to execute:

a speech recognition step of checking the uttered speech against registered words stored in a speech recognition word database which defines vocabulary for speech recognition and stores the vocabulary as registered words;

a reference similarity calculation step of finding a similarity between the uttered speech and a combination of acoustic units, which are subwords;

an unregistered word judgment step of judging, based on a result of the check by said speech recognition unit and a result of the calculation performed by said reference similarity calculation unit, whether the uttered speech is a registered word which is stored in the speech recognition word database or an unregistered word which is not stored in the speech recognition word database;

an unregistered word candidate search step of searching for an unregistered word candidate that is likely to correspond to the uttered speech, based on the uttered speech, among unregistered words stored in an unregistered word database, in the case where the uttered speech is judged to be an unregistered word in said unregistered word judgment step; and

a result display step of displaying a result of the search.

18. A program used for a speech recognition device that recognizes uttered speech, said program causing a computer to execute:

a speech recognition step of checking the uttered speech against registered words stored in a speech recognition word database which defines vocabulary for speech recognition and stores the vocabulary as registered words;

an unregistered word judgment step of judging, based on a result of the check in said speech recognition step, whether the uttered speech is a registered word which is stored in the speech recognition word database or an unregistered word which is not stored in the speech recognition word database;

an unregistered word candidate search step of searching for an unregistered word candidate that is likely to correspond to the uttered speech, based on the uttered speech, among unregistered words stored in an unregistered word database, in the case where the uttered speech is judged to be an unregistered word in said unregistered word judgment step; and

a result display step of displaying a result of the search.

20. A speech recognition method for recognizing uttered speech, comprising:

a speech recognition step of checking the uttered speech against registered words stored in a speech recognition word database which defines vocabulary for speech recognition and stores the vocabulary as registered words;

an unregistered word judgment step of judging, based on a result of the check in said speech recognition step, whether the uttered speech is a registered word which is stored in the speech recognition word database or an unregistered word which is not stored in the speech recognition word database;

an unregistered word candidate search step of searching for an unregistered word candidate that is likely to correspond to the uttered speech, based on the uttered speech, among unregistered words stored in an unregistered word database, in the case where the uttered speech is judged to be an unregistered word in said unregistered word judgment step; and

a result display step of displaying a result of the search.