Automated retrieval of voice mail using speech recognition

• US 5,822,405 A
• Filed: 09/16/1996
• Issued: 10/13/1998
• Est. Priority Date: 09/16/1996
• Status: Expired due to Term

• Current Assignee: Toshiba Corporation
• Original Assignee: Toshiba America Information Systems Incorporated (Toshiba Corporation)
• Inventors: Astarabadi, Shaun
• Primary Examiner(s): Tsang, Fan S.
• Application Number:
  US08/710,366
• Time in Patent Office: 757 Days
• Field of Search: 379/67, 379/88, 379/89, 379/92.01, 379/92.03, 379/92.04, 379/93.24, 379/352
• US Class Current: 379/88.04
• CPC Class Codes: H04M 1/271 controlled by voice recognition
• CPC Class Codes: H04M 1/6505 storing speech in digital form
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 2203/253 where a visual mode is used instead of a voice mode
• CPC Class Codes: H04M 3/533 Voice mail systems
• CPC Class Codes: H04M 3/53333 Message receiving aspects
• CPC Class Codes: H04Q 3/625 Arrangements in the private branch exchange

Abstract

In accordance with the invention, a traveler can use a portable computer to automatically retrieve voice mail from a PBX-based voice mail system or a telephone answering machine. The computer has a modem, an audio output system (e.g., SoundBlaster card and speakers) and a special program. During a training phase of the method, the program creates a script of normal interaction with the voice mail system. The script is used during a message retrieval phase to retrieve voice mail messages. Multiple scripts may be recorded. During the retrieval phase of the method, the program uses word spotting to compare prompts received from the voice mail system to prompts in the script to ensure that the process is working properly. The program can create script commands for: dialing the voice mail system; entering a mail box number and password; making appropriate selections to listen to new messages; recording the messages from the voice mail system into the portable computer; and disconnecting from the voice mail system.

Claims

1. A computer device for accessing and processing at least one audio message from a first remote message facility accessible via telephone, where the first remote message facility provides verbal announcements, the computer device comprising:

a processor;

a memory coupled to the processor for storing a program, verbal announcement audio signals, and input message signals;

a telephone interface coupled to the processor for selectively and temporarily interconnecting the computer device with a telephone network;

an audio input subsystem coupled to the processor for accepting the verbal announcement audio signals and the input message signals;

an audio output subsystem coupled to the processor for outputting stored input message signals; and

a signal producing device for producing signals recognizable by the first remote message facility;

wherein the program comprises;

a verbal announcement accepting function for accepting verbal announcements from the first remote message facility,

a verbal announcement recording function for recording verbal announcements from the first remote message facility,

a verbal announcement discriminating function for discriminating between different verbal announcements from the first remote message facility, and

a response function for responding to verbal announcements from the first remote message facility.

2. A computer device for accessing and processing audio messages stored at a first remote message facility accessible via telephone according to claim 1 wherein the computer device is a portable computer and the processor is a microprocessor.

3. A portable computer device for accessing and processing audio messages stored at a first remote message facility accessible via telephone as set forth in claim 2 wherein the input message signals are audible speech signals.

4. A portable computer device for accessing and processing audio messages stored at a first remote message facility accessible via telephone as set forth in claim 3 further including an analog-to-digital converter coupled to the telephone interface for converting audible speech messages transferred from the first remote message facility into digital data format prior to storage within the memory.

5. A portable computer device for accessing and processing audio messages stored at a first remote message facility accessible via telephone as set forth in claim 3 wherein the audio output subsystem further includes headphones for monitoring verbal announcement signals.

6. A portable computer device for accessing and processing audio messages stored at a first remote message facility accessible via telephone as set forth in claim 2 wherein the verbal announcement discriminating function further comprises a spotting function for spotting a word or phrase within a continuous flow of speech.

7. A portable computer device for processing messages stored at a first remote message facility accessible via telephone as set forth in claim 2 wherein the response function outputs speech signals.

8. A computer device as set forth in claim 1 wherein the computer device also functions to access and process at least one audio message from a second remote message facility, where the second remote message facility provides different verbal announcements from the first remote message facility.

9. A computer device as set forth in claim 8 wherein the computer device automatically accesses the second remote message facility upon recognizing an error condition affecting its access to the first remote message facility.

46. A computer device as set forth in claim 1 wherein the computer device also functions to access and process at least one audio message from a second remote message facility, where the second remote message facility is an answering machine.

10. A portable computer device for accessing and processing audio messages stored at a first remote message facility accessible via telephone, where the first remote message facility provides verbal announcements, the portable computer device comprising:

a microprocessor;

memory coupled to the microprocessor for storing a program and audio signals;

a telephone interface coupled to the microprocessor for selectively and temporarily interconnecting the portable computer device with a telephone network;

a sound board coupled to the microprocessor for accepting audio signals;

an audio output coupled to the microprocessor for outputting audible speech from stored audio signals; and

a signal producing device for producing signals recognizable by the first remote message facility;

wherein the program comprises;

a verbal announcement accepting function for accepting verbal announcements from the first remote message facility,

a verbal announcement recording function for recording verbal announcements from the first remote message facility,

a verbal announcement discriminating function for recognizing verbal announcements from the first remote message facility,

an output control function for allowing a user to control a message playback with respect to identity of message and speed of playback, and

a response function for responding to verbal announcements from the first remote message facility.

11. A portable computer device for accessing and processing audio messages stored at a first remote message facility accessible via telephone as set forth in claim 10 wherein the audio signals stored within the memory are stored in digital data format and wherein the audio output includes digital data-to-speech conversion means coupled to the microprocessor for creating speech signals corresponding to digital data.

12. A portable computer device for processing messages stored at a first remote message facility accessible via telephone as set forth in claim 10 further including analog storage means and wherein a set of the audio signals are stored in analog data format.

13. A portable computer device for processing messages stored at a first remote message facility accessible via telephone as set forth in claim 10 further comprising an analog-to-digital converter coupled to the sound board for converting audio messages transferred from the first remote message facility into digital data format prior to storage within the memory.

14. A portable computer device for processing messages stored at a first remote message facility accessible via telephone as set forth in claim 10 further including a status indicator coupled to the microprocessor for visually indicating communication between the first remote message facility and the portable computer device.

15. A portable computer device for processing messages stored at a first remote message facility accessible via telephone as set forth in claim 10 wherein the response function outputs DTMF signals.

16. A portable computer device for processing messages stored at a first remote message facility accessible via telephone as set forth in claim 10 wherein the response function outputs speech signals.

17. A portable computer device as set forth in claim 10 wherein the portable computer device also functions to access and process at least one audio message from a second remote message facility, where the second remote message facility provides different verbal announcements from the first remote message facility.

18. A portable computer device as set forth in claim 17 wherein the portable computer device automatically accesses the second remote message facility upon recognizing an error condition affecting its access to the first remote message facility.

19. A computer device for accessing and processing audio messages stored at more than one remote message facility accessible via telephone, where each remote message facility provides verbal announcements, the computer device comprising:

a processor;

memory coupled to the processor for storing a program and audio signals;

a telephone interface coupled to the processor for interconnecting the computer device with a telephone network;

a sound board coupled to the processor for accepting audio signals;

an audio output coupled to the processor for outputting audible speech from stored audio signals; and

a signal producing device for producing signals recognizable by each remote message facility;

wherein the program comprises;

a verbal announcement accepting function for accepting verbal announcements from each remote message facility,

a verbal announcement recording function for recording verbal announcements from each remote message facility,

a verbal announcement discriminating function for recognizing verbal announcements from each remote message facility, and

a response function for responding appropriately to verbal announcements from each remote message facility.

20. A computer device for accessing and processing audio messages stored at more than one remote message facility accessible via telephone, in accordance with claim 19, wherein the computer is a portable computer and the processor is a microprocessor.

21. A computer device for accessing and processing audio messages stored at more than one remote message facility accessible via telephone, in accordance with claim 20, wherein a first remote message facility is a voice mail facility and a second remote message facility is an answering machine.

22. A computer device for accessing and processing audio messages stored at more than one remote message facility accessible via telephone, in accordance with claim 20, wherein the stored program further comprises an output control function for allowing a user to control a message playback as to identity of message and speed of playback.

23. A computer device for accessing and processing at least one audio message from a first remote message facility accessible via telephone, where the first remote message facility provides verbal announcements, the computer device comprising:

a processor;

memory coupled to the processor for storing a program, verbal announcement audio signals, and input message signals;

a telephone interface coupled to the microprocessor for selectively and temporarily interconnecting the computer device with a telephone network;

a sound board coupled to the processor for accepting the verbal announcement audio signals and the input message signals;

an audio output coupled to the processor for outputting stored input message signals; and

a signal producing device for producing signals recognizable by the first remote message facility;

wherein the stored program is for performing the steps of;

(1) causing the computer device to connect via a telephone line, to the first remote message facility;

(2) causing the computer device to accept a first voice greeting announcement from the first remote message facility to the computer device;

(3) causing the computer device to compare the first voice greeting announcement with a second voice greeting announcement previously stored in the computer device;

(4) causing the computer device to test if the first voice greeting announcement conforms with the second voice greeting announcement, and if so then;

(a) causing the computer device to receive a first voice status announcement from the first remote message facility, wherein the first voice status announcement indicates that the first remote message facility has unread messages;

(b) causing the computer device to compare the first voice status announcement with a second voice status announcement previously stored in the computer device;

(c) causing the computer device to test if the first voice status announcement conforms with the second voice status announcement, and if so then

(i) causing the computer device to accept from the first remote message facility first set of unread messages; and

(ii) saving the first set of unread messages on the computer device.

24. A computer device for accessing and processing audio messages stored at a first remote message facility accessible via telephone as set forth in claim 23 wherein the stored program is for performing the further steps of:

causing the computer device to accept a mailbox identification request announcement from the first remote audio message facility; and

causing the computer device to transmit a mailbox identification number, previously stored in the computer device, from the computer device to the first remote message facility.

25. A computer device for accessing and processing audio messages stored at a first remote message facility accessible via telephone as set forth in claim 24 wherein the stored program is for performing the further steps of:

causing the computer device to test if the first remote message facility indicates that the mailbox identification number matches the authorized mailbox number, then

causing the computer device to transmit a first password, previously stored in the computer, from the computer to the first remote message facility.

26. A computer device for accessing and processing at least one audio message from a first remote message facility as set forth in claim 23, wherein the computer device also functions to access and process at least one audio message from a second remote message facility and the stored program is for performing the additional steps of:

(4) causing the computer device to connect via the telephone line, to the second remote message facility;

(5) causing the computer device to accept a third voice greeting announcement from the second remote message facility to the computer device;

(6) causing the computer device to compare the third voice greeting announcement with a fourth voice greeting announcement previously stored in the computer device;

(7) causing the computer device to test if the third voice greeting announcement conforms with the fourth voice greeting announcement, and if so then

(a) causing the computer device to accept from the first remote message facility a second set of unread messages; and

(b) causing the computer device to save the second set of unread messages.

27. A computer device for accessing and processing at least one audio message from a first remote message facility as set forth in claim 26, wherein the stored program is farther for performing the additional step of:

(8) causing the computer device to determine if there is an error condition with respect to the first remote message facility and, if there is an error condition with respect to the first message facility, and if so then;

(a) causing the computer device to disconnect from the first remote message facility, and

(b) causing the computer device to proceed to establish a connection with the second remote message facility.

28. A method of retrieving messages from a remote audio message facility where the remote audio message facility provides verbal announcements, comprising the steps of:

(1) connecting a computer via a telephone to the remote message facility;

(2) accepting a first voice greeting announcement, previously stored in the remote message facility, from the remote message facility to the computer;

(3) comparing in the computer the first greeting announcement with a second voice greeting announcement previously stored in the computer;

(4) if the first greeting announcement conforms with the second greeting announcement, then;

(a) transmitting a first password, previously stored in the computer, from the computer to the remote message facility;

(b) comparing in the remote message facility the first password with a second password previously stored in the remote message facility;

(c) if the first password matches the second password, then

(i) receiving a first voice status announcement in the computer from the remote message facility, wherein the first voice status announcement indicates that the remote message facility has unread messages;

(ii) comparing in the computer the first voice status announcement with a second voice status announcement previously stored in the computer;

(iii) if the first voice status announcement conforms with the second voice status announcement, then accepting transfer, from the remote message facility to the computer, of the unread messages.

29. A method of retrieving messages from a remote message facility as set forth in claim 28, wherein the step of accepting transfer from the remote message facility to the computer of the unread messages comprises the steps of:

(1) transmitting an unread message from the remote message facility to the computer;

(2) identifying in the computer the end of transmission of the unread message;

(3) transmitting a third voice status announcement, previously stored in the remote message facility, from the remote message facility to the computer;

(4) comparing in the computer the third voice status announcement with a fourth voice status announcement previously stored in the personal computer, the fourth voice status announcement indicative that the remote message facility has additional unread messages;

(5) if the third voice status announcement conforms with the fourth voice status announcement, then returning to said step of transmitting an unread message above.

30. A method of retrieving messages from a remote message facility as set forth in claim 28, wherein the step of accepting transfer from the remote message facility to the computer of the unread messages comprises the steps of:

(1) transmitting an unread message from the remote message facility to the computer;

(2) identifying in the computer the end of transmission of the unread message;

(3) transmitting a current DTMF signal, previously stored in the remote message facility, from the remote message facility to the computer;

(4) comparing in the computer the current DTMF signal with a previously stored DTMF signal in the personal computer, the stored DTMF signal indicative that the remote message facility has additional unread messages;

(5) if the current DTMF signal conforms with the stored DTMF signal, then returning to said step of transmitting an unread message above.

31. A method of retrieving at least one audio message from a remote audio message facility accessible via telephone, where the remote message facility provides verbal announcements, the method comprising the steps of:

(1) connecting a computer containing the computer program, via a telephone line, to the remote message facility;

(2) accepting a first voice greeting announcement from the remote message facility to the computer;

(3) comparing in the computer the first voice greeting announcement with a second voice greeting announcement previously stored in the computer;

if the first voice greeting announcement conforms with the second voice greeting announcement, then;

(a) receiving a first voice status announcement by the computer from the remote message facility, wherein the first voice status announcement indicates that the remote message facility has unread messages;

(b) comparing in the computer the first voice status announcement with a second voice status announcement previously stored in the computer;

if the first voice status announcement conforms with the second voice status announcement, then

(i) accepting from the remote message facility to the computer unread messages; and

(ii) saving the unread messages on the computer.

32. A computer device for processing verbal communications obtained from interaction with a first remote facility accessible via telephone, the computer device comprising:

a processor means;

a data input means for interacting with the processor means;

a display means for providing output to a user;

a communication input means for sampling verbal communications from the first remote facility;

a memory means coupled to the processor means for storing a plurality of messages;

a telephone interface means coupled to the processor for interconnecting the computer device with a telephone network;

communications means coupled to the processor means for transferring at least one audio input from the first remote message facility to the memory in response to establishing communication between the first remote message facility and the computer device;

first comparison means for comparing a first announcement received from the first remote message facility with a second announcement received from the first remote message facility and stored in the memory to determine if the first announcement matches the second announcement; and

first control means for actuating receipt and transfer of an audio message from the first remote facility to the memory when the first announcement matches the second announcement.

33. A computer device for processing verbal communications obtained from interaction with a first remote facility accessible via telephone as set forth in claim 32 which further includes audio output means for creating audible speech from an audio message stored in the computer memory.

34. A computer device for processing verbal communications obtained from interaction with a first remote facility accessible via telephone as set forth in claim 33 further including an analog-to-digital converter coupled to the communications means for converting audio messages transferred from the first remote facility into digital data format prior to storage within the memory means.

35. A computer device for processing verbal communications obtained from interaction with a first remote facility accessible via telephone as set forth in claim 34 wherein the first remote facility is a voice mail facility containing stored audio messages.

36. A computer device for processing verbal communications obtained from interaction with a first remote facility accessible via telephone as set forth in claim 32 and also for processing verbal communications from interaction with a second remote facility accessible via telephone, further comprising:

second comparison means for comparing a third announcement received from the second remote message facility with a fourth announcement received from the second remote message facility and stored in the memory to determine if the third announcement matches the fourth announcement; and

second control means for actuating receipt and transfer of an audio message from the second remote facility to the memory when the third announcement matches the fourth announcement.

37. A computer device for processing verbal communications obtained from interaction with a first remote facility and a second remote facility, each accessible via telephone as set forth in claim 36, further comprising:

recognition means for recognizing an error condition with respect to the first remote facility; and

error handling means for terminating connection with the first remote facility and instituting connection with the second remote facility in the event of an error condition with respect to the first remote facility.

38. A computer device for processing verbal communications obtained from interaction with a first remote facility accessible via telephone as set forth in claim 37 wherein the first remote facility is a voice mail facility capable of containing stored audio messages and the second remote facility is an answering machine.

39. A method of training a computer to provide automatic communications with a remote message facility where the remote message facility provides verbal announcements for use by people to navigate features provided by the remote message facility the method comprising the steps of:

(1) running an access program on the computer;

(2) connecting the computer running the access program to the remote message facility; and

(3) navigating a verbal communication accessing session generated by the remote message facility by;

(a) the computer capturing at least one verbal announcement from the remote message facility,

(b) the computer analyzing and reproducing the verbal announcement from the remote message facility for the user to hear, the verbal announcement including a key phrase,

(c) the computer storing an identifiable portion of the key phrase of the verbal announcement,

(d) the computer receiving input from the user and transmitting the input to the remote message facility, and

(e) the computer capturing the user input.

40. A method of training a computer to provide automatic communications with a remote message facility as set forth in claim 39 wherein the remote facility is a voice mail facility.

41. A method of training a computer to provide automatic communications with a remote message facility as set forth in claim 39 wherein the remote facility is an answering machine.

42. A method of training a computer to provide automatic communications with a remote message facility as set forth in claim 39 wherein the user input is in the form of DTMF.

43. A method of training a computer to provide automatic communications with a remote message facility as set forth in claim 39 wherein the user input is verbal.

44. A method of training a computer to provide automatic communications with a remote message facility as set forth in claim 39 where the communications from the remote facility are captured as wave records linked to corresponding user input.

45. A method of training a computer to provide automatic verbal communications with a remote message facility where the remote message facility provides verbal announcements comprising the steps of:

(1) running an access program on a personal computer;

(2) connecting the computer running the access program to the remote message facility;

(3) setting up a user interface to allow interaction of a user with the remote facility;

(4) navigating a verbal communication accessing session with respect to the remote facility by;

(a) allowing the user to perceive communications from the remote facility,

(b)allowing the user to transmit input to the remote facility,

(c) capturing the communications from the remote facility, and

(d) capturing the user input; and

(5) processing the communications from the remote facility to produce recognition contexts,

where the communications from the remote facility are captured as wave records

linked to corresponding user input; and

where the step of processing the communications from the remote facility further comprises the steps of;

(a) converting the wave records to wave format files;

(b) pre-processing the wave format files;

(c) editing the pre-processed wave format files with a utility program; and

(d) evaluating the edited wave format files with an automatic speech recognition evaluator.