Methods and apparatus for performing speaker independent recognition of commands in parallel with speaker dependent recognition of names, words or phrases

• US RE38,101 E1
• Filed: 02/16/2000
• Issued: 04/29/2003
• Est. Priority Date: 02/29/1996
• Status: Expired due to Term

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Telesector Resources Group Incorporated
• Inventors: Lubensky, David M., Asadi, Ayman O., Raman, Vijay R., Naik, Jayant M., Vysotsky, George J.
• Primary Examiner(s): Hong, Harry S.
• Application Number:
  US09/505,103
• Time in Patent Office: 1,168 Days
• Field of Search: 379/67.1, 379/79, 379/84, 379/88.01, 379/88.03, 379/88.28, 379/188, 379/189, 379/199, 379/201.01, 379/216.01, 379/361, 704/246, 704/247, 704/275, 704/272, 704/201, 704/211, 704/270.1
• US Class Current: 379/88.03
• CPC Class Codes: G10L 15/065 Adaptation
• CPC Class Codes: G10L 15/20 Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech G10L21/02 takes precedence
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 15/34 Adaptation of a single recogniser for parallel processing, e.g. by use of multiple processors or cloud computing
• CPC Class Codes: G10L 2015/088 Word spotting
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: H04M 1/271 controlled by voice recognition
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 3/42 Systems providing special services or facilities to subscribers specially adapted for wireless communication networks H04W4/00
• CPC Class Codes: H04M 3/42204 Arrangements at the exchange for service or number selection by voice at the terminal H04M1/27
• CPC Class Codes: H04M 3/44 Additional connecting arrangements for providing access to frequently-wanted subscribers, e.g. abbreviated dialling at the subscriber's set H04M1/27; automatic redialling H04M3/424

Abstract

Methods and apparatus for activating telephone services in response to speech are described. A directory including names is maintained for each customer. A speaker dependent speech template and a telephone number for each name, is maintained as part of each customer'"'"'s directory. Speaker independent speech templates are used for recognizing commands. The present invention has the advantage of permitting a customer to place a call by speaking a person'"'"'s name which serves as a destination identifier without having to speak an additional command or steering word to place the call. This is achieved by treating the receipt of a spoken name in the absence of a command as an implicit command to place a call. Explicit speaker independent commands are used to invoke features or services other than call placement. Speaker independent and speaker dependent speech recognition are performed on a customer'"'"'s speech in parallel. An arbiter is used to decide which function or service should be performed when an apparent conflict arises as a result of both the speaker dependent and speaker independent speech recognition step outputs. Stochastic grammars, word spotting and/or out-of-vocabulary rejection are used as part of the speech recognition process to provide a user friendly interface which permits the use of spontaneous speech. Voice verification is performed on a selective basis where security is of concern.

Claims

1. A method of providing a service in response to speech, comprising the steps of:

identifying the speaker;

performing in parallel,

i. a speaker independent speech recognition operation to identify a spoken command;

ii. a speaker dependent speech recognition operation in an attempt to identify a word; and

performming an operation in response to the spoken command identified by performing the speaker independent speech recognition operation.

2. The method of claim 1,

wherein the spoken command is a command to dial a telephone number; and

wherein the step of performing an operation includes the step of;

dialing a telephone number associated with a word detected by performing the speaker dependent speech recognition operation.

3. The method of claim 2, wherein the word is a name, the method further comprising the steps of:

retrieving from a database, a plurality of speaker dependent speech templates associated with the identified speaker to be used when performing the speaker dependent speech recognition operation; and

retrieving from a data base, a telephone number associated with the name identified by performing the speaker dependent speech recognition operation.

4. The method of claim 2, further comprising the step of:

retrieving from a database, a plurality of speaker dependent speech templates associated with the identified speaker, to be used when performing the speaker dependent speech recognition operation.

5. The method of claim 2, further comprising the step of:

performing an arbitration operation to determine whether the output of the speaker independent speech recognition operation or the output of the speaker dependent speech recognition operation should be used when there is a conflict between the output of the two speech recognition operations.

6. A method of providing a telephone service in response to a caller'"'"'s speech, the method comprising the steps of:

identifying the caller;

performing, in parallel,

i. a speech recognition operation on the speech to identify an explicit command in the speech;

ii. a speaker dependent speech recognition operation on the speech to identify a word, other than an explicit command, in the speech; and

performing an action as a function of the outcome of the speech recognition operations performed in parallel.

7. The method of claim 6,

wherein the word is a name; and

wherein the step of performing an action includes the step of;

dialing a telephone number associated with the name.

8. The method of claim 6, further comprising the steps of:

detecting a first speech time interval to which an identified explicit command corresponds;

detecting a second speech time interval to which an identified word corresponds; and

if there is a substantial overlap between the first and second time intervals, performing an arbitration operation to determine whether to respond to the detected command or to take some other action.

9. The method of claim 8, wherein a substantial overlap between the first and second time intervals exists when the first and second time intervals overlap by 50% or more.

10. A device for responding to speech, comprising:

means for performing speaker independent speech recognition on the speech to detect the presence of a spoken command in the speech;

means for performing speaker dependent speech recognition on the speech to detect a non-command word in the speech, the speaker independent and speaker dependent speech recognition means operating in parallel; and

a device for performing an action in response to the detection of a command by the speaker independent speech recognition means.

11. The device of claim 10, further comprising:

an arbiter coupled to the speaker independent and speaker dependent speech recognition means, the arbiter for determining the action to be taken in response to the detection of a spoken command by the speaker independent speech recognition means and a non-command word by the speaker dependent speech recognition means.

12. The device of claim 11, further comprising a voice verification circuit coupled to the arbiter for selectively performing voice verification on the speech.

13. The device of claim 12, further comprising:

a database coupled to the speaker dependent speech recognition means, the database including a plurality of speaker dependent speech recognition templates and telephone numbers, a telephone number being associated with each speaker dependent speech recognition template.

14. The device of claim 13, further comprising:

a telephone for receiving the speech from a caller;

a switch for coupling the telephone to the speaker independent and speaker dependent speech recognition means.

15. A method of providing a telephone service in response to speech of a caller, the method comprising the steps of:

16. The method of claim 15, further comprising the steps of:

17. The method of claim 16, wherein the second speech recognition operation is a speaker independent speech recognition operation.

18. The method of claim 17, wherein the first speech recognition operation is a speaker dependent speech recognition operation.

19. The method of claim 18, wherein the first and second speech recognition operations are performed in parallel.

20. The method of claim 19, further comprising the step of:

21. The method of claim 19,

22. A voice dialing system which is responsive to the speech of a system user, the system comprising:

23. The voice dialing system of claim 22, further comprising:

24. The voice dialing system of claim 23, wherein means for performing a first speech recognition operation and means for performing a second speech recognition operation are arranged in parallel with one another.

25. The voice dialing system of claim 24, wherein the means for performing a second speech recognition operation is a speaker independent speech recognizer.

26. The voice dialing system of claim 25, wherein the means for performing a first speech recognition operation is a speaker dependent speech recognizer.