Methods and apparatus for activating telephone services in response to speech

• US 5,719,921 A
• Filed: 02/29/1996
• Issued: 02/17/1998
• Est. Priority Date: 02/29/1996
• Status: Expired due to Term

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Nynex Science & Technology
• Inventors: Naik, Jayant M., Raman, Vijay R., Lubensky, David M., Asadi, Ayman O., Vysotsky, George J.
• Primary Examiner(s): Hong, Harry S.
• Application Number:
  US08/609,029
• Time in Patent Office: 719 Days
• Field of Search: 379/67, 379/88, 379/89, 379/201, 379/189, 379/199, 379/188, 379/91, 379/79, 379/84, 379/216, 379/361, 381/41, 381/42, 381/43, 395/2.55, 395/2.56, 395/2.84, 395/2.81, 395/2, 395/2.1, 395/2.2, 395/2.79
• US Class Current: 379/88.01
• 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 telephone service in response to speech of a caller, the method comprising the steps of:

identifying the caller;

accessing a database including speaker dependent speech recognition templates for a plurality of names;

performing a speaker dependent speech recognition operation on the speech using the speaker dependent speech recognition templates to detect a presence of one of the plurality of names in the speech;

performing speech recognition on the speech in order to detect a presence of an explicit command in the speech; and

upon detecting the presence of one of the plurality of names and failing to detect the presence of an explicit command in the speech, dialing a telephone number corresponding to the one of the plurality of names whose presence was detected.

2. The method of claim 1, wherein the step of performing speech recognition on the speech in order to detect the presence of an explicit command includes the step of:

performing a speaker independent speech recognition operation on the speech.

3. The method of claim 2, wherein the speaker dependent speech recognition operation and the speaker independent speech recognition operation are performed in parallel.

4. The method of claim 3, wherein the explicit command may be a command of a first type which does not require the additional input of a name to be performed, the method further comprising the step of:

upon detecting the presence of one of the plurality of names as a result of the speaker dependent speech recognition operation and failing to detect a command of the first type;

performing an arbitration step to determine whether to respond to the detected command or to dial the telephone number corresponding to the detected one of the plurality of names.

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

generating a first confidence measure when performing the speaker dependent speech recognition operation and the presence of one of the plurality of names is detected, the first confidence measure being an indication of how likely the result of the speaker dependent speech recognition operation is correct;

generating a second confidence measure when performing the speaker independent speech recognition operation and the explicit command is detected, the second confidence measure being an indication of how likely the result of the speaker independent speech recognition operation is correct; and

using the first and second confidence measures to perform the arbitration step.

6. The method of claim 5, wherein the arbitration step is performed as a function of durational models of names and commands and the statistical probability of the occurrence of the names and the commands.

7. The method of claim 1, wherein the explicit command may be any one of a plurality of different types of explicit commands, and where the method further comprises the step of:

performing a voice verification operation to confirm the identity of the caller upon the detection of a preselected type of explicit command.

8. The method of claim 7, wherein the preselected type of explicit command involves a financial transaction.

9. The method of claim 7, wherein the preselected type of explicit command is a request for access to stored information.

10. The method of claim 3, wherein the explicit command whose presence is detected may be of a second type which requires the additional input of a name to be performed, the method further comprising the steps of:

determining if the explicit command whose presence was detected is of the second type; and

if the explicit command whose presence was detected is of the second type;

determining if the required name was detected by the speaker dependent speech recognition process;

if the required name was detected, performing an operation in response to the explicit command whose presence was detected and the detected name; and

if the required name was not detected, prompting the caller for additional input.

11. The method of claim 10, wherein the second type of command is a call forwarding command.

12. A method of responding to speech from a caller, the method comprising the steps of:

performing a speaker dependent speech recognition operation on the speech to detect a presence of a name in the speech;

performing a speaker independent speech recognition operation on the speech in order to detect a presence of an explicit command in the speech, the speaker dependent and speaker independent speech recognition operations being performed in parallel;

upon detecting the presence of a name and failing to detect the presence of an explicit command in the speech;

dialing a telephone number corresponding to the name whose presence is detected; and

upon detecting the presence of a name and the presence of an explicit command, performing an arbitration step to determine whether to respond to the detected command or to dial the telephone number corresponding to the name whose presence is detected if the explicit command whose presence was detected is of a first type which does not require a name to be performed.

13. The method of claim 12, further comprising the step of:

upon detecting the presence of a command of a second type;

performing a voice verification operation to check the identity of the caller.

14. The method of claim 13, wherein speaker dependent voice recognition and speaker independent voice recognition are performed using word spotting and out of vocabulary word rejection techniques.

15. A device for responding to a caller'"'"'s speech, comprising:

means for performing speaker dependent speech recognition on the speech to detect the presence of a name in the speech;

means for performing speaker independent speech recognition on the speech to detect the presence of an explicit command in the speech, the means for performing speaker dependent and speaker independent speech recognition being coupled in parallel; and

means for dialing a telephone number corresponding to the name whose presence is detected, upon detecting the presence of a name and failing to detect the presence of an explicit command in the speech, coupled to the means for performing speaker independent and speaker dependent speech recognition.

16. The device of claim 15, further comprising:

an arbiter coupled to the means for performing speaker dependent and speaker independent speech recognition for receiving an output therefrom and for determining the action to be taken in response to the detection of names and commands.

17. The device of claim 16, further comprising a voice verification circuit coupled to the arbiter for selectively performing voice verification on the speech.

18. The device of claim 17, further comprising:

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

19. The device of claim 18 further comprising a switch coupled to the caller and to the means for performing speaker independent and speaker dependent speech recognition.

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

identifying the caller;

accessing a database including speaker dependent speech recognition templates of names;

performing speaker dependent speech recognition on the speech using the speaker dependent speech recognition templates to detect a name in the speech and a first time interval in which the name was spoken;

performing speech recognition on the speech in order to detect an explicit command in the speech and a second time interval in which the explicit command was spoken; and

upon detecting the name and the explicit command determining if the explicit command is of a type which requires a name to be performed and determining if there is a significant overlap between the first and second time intervals;

if it is determined that the command is of the type which requires a name to be performed and if the first and second time intervals significantly overlap, performing an arbitration step to determine whether to respond to the detected command or to dial a telephone number in response to the detected name.

21. The method of claim 20,

wherein the step of performing speech recognition in order to detect the presence of an explicit command includes the step of performing speaker independent speech recognition;

and

wherein the steps of performing speaker dependent speech recognition and speaker independent speech recognition are performed in parallel.

22. The method of claim 21, wherein upon detecting the presence of a name and failing to detect the presence of an explicit command in the speech a further step of dialing a telephone number, corresponding to the name whose presence was detected, is performed.