Computer-based streaming voice data contact information extraction

• US 9,936,068 B2
• Filed: 08/04/2014
• Issued: 04/03/2018
• Est. Priority Date: 08/04/2014
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Kau, Christian B., Schoudt, Jerald T.
• Primary Examiner(s): BEZUAYEHU, SOLOMON G
• Application Number:
  US14/450,411
• Publication Number:
  US 20160036969A1
• Time in Patent Office: 1,338 Days
• Field of Search: 379 8801- 8819
• CPC Class Codes: H04M 3/42093 Notifying the calling party of information on the called or connected party
• CPC Class Codes: H04M 3/42221 Conversation recording systems at the subscriber's set H04M1/656

Abstract

Embodiments relate to extracting contact information from streaming voice data. An aspect includes a speech recognition module configured to transcribe a stream of voice data representing at least a portion of a telephone conversation into text data; a contact information extraction module configured to extract contact information from the text data; and a transceiver configured to send at least a portion of the contact information to a recipient calling device.

Claims

1. A method for extracting contact information from streaming voice data, the method comprising:

transcribing, into a string of text data, a stream of voice data representing a portion of a telephone conversation between a caller at an origination calling device and a call recipient at a recipient calling device, the transcribing performed automatically by a processor based on a determination that a caller identifier (ID) of the origination calling device does not match an entry in an address book located on the recipient calling device;

identifying, by the processor, at least one key word or key phrase in the text data;

extracting contact information from the text data, the extracting by the processor and including extracting information preceding or following the identified key word or key phrase as at least a portion of the contact information;

determining whether the caller is identified in the contact information by a first name only or by a full name;

determining a probability that the caller is associated with an entry in the address book based at least in part on whether the caller is identified in the contact information by a first name only or by a full name, wherein a caller identified by a first name only has a higher probability of being associated with an entry in the address book than a caller identified by a full name;

performing the transcribing for an amount of time that is based at least in part on the probability that the caller is associated with an entry in the address book, wherein the amount of time decreases as the probability increases; and

prompting the call recipient to store at least a portion of the contact information to the recipient calling device, the prompting based at least in part on the probability that the caller is associated with an entry in the address book being below a threshold.

2. The method of claim 1, wherein the stream of voice data is transcribed during the telephone conversation between the caller and the call recipient.

3. The method of claim 1, wherein the stream of voice data is recorded during the telephone conversation between the caller and the call recipient.

4. The method of claim 1, wherein the transcribing comprises applying speech recognition technology.

5. The method of claim 1, further comprising identifying a phrase associated with the caller in the text data.

6. The method of claim 1, wherein the storing is in the address book.

17. The method of claim 6, wherein the storing is during the telephone conversation.

7. A system for extracting contact information from streaming voice data, the system comprising:

a memory having computer readable computer instructions; and

a processor for executing the computer readable instructions, the computer readable instructions including;

transcribing, into a string of text data, a stream of voice data representing a portion of a telephone conversation between a caller at an origination calling device and a call recipient at a recipient calling device and send contact information to the recipient calling device, the transcribing performed automatically based on a determination that a caller identifier (ID) of the origination calling device does not match an entry in an address book located on the recipient calling device;

identifying at least one key word or key phrase in the text data;

extracting the contact information from the text data, the extracting including extracting information preceding or following the identified key word or key phrase as at least a portion of the contact information;

determining whether the caller is identified in the contact information by a first name only or by a full name;

determining a probability that the caller is associated with an entry in the address book based at least in part on whether the caller is identified in the contact information by a first name only or by a full name, wherein a caller identified by a first name only has a higher probability of being associated with an entry in the address book than a caller identified by a full name;

performing the transcribing for an amount of time that is based at least in part on the probability that the caller is associated with an entry in the address book, wherein the amount of time decreases as the probability increases; and

prompting the call recipient to store at least a portion of the contact information to the recipient calling device, the prompting based at least in part on the probability that the caller is associated with an entry in the address book being below a threshold.

8. The system of claim 7, wherein the stream of voice data is transcribed during the telephone conversation between the caller and the call recipient.

9. The system of claim 7, further comprising a computational linguistics module configured to analyze the text data to determine the meaning of a word.

10. The system of claim 7, further comprising a key phrase identification module configured to identify a phrase associated with the caller in the text data.

11. The system of claim 7, wherein the stream of voice data is transcribed during the telephone conversation between the caller and the call recipient.

12. The system of claim 7, wherein the storing is in the address book.

18. The system of claim 12, wherein the storing is during the telephone conversation.

13. A computer program product for extracting contact information from streaming voice data, the computer program product comprising:

a non-transitory computer readable storage medium having program code embodied therewith, the program code executable by a computer to implement;

transcribing, into a string of text data, a stream of voice data representing a portion of a telephone conversation between a caller at an origination calling device and a call recipient at a recipient calling device, the transcribing performed automatically based on a determination that a caller identifier (ID) of the origination calling device does not match an entry in an address book located on the recipient calling device;

identifying at least one key word or key phrase in the text data;

extracting contact information from the text data, the extracting including extracting information preceding or following the identified key word or key phrase as at least a portion of the contact information;

determining whether the caller is identified in the contact information by a first name only or by a full name;

determining a probability that the caller is associated with an entry in the address book based at least in part on whether the caller is identified in the contact information by a first name only or by a full name, wherein a caller identified by a first name only has a higher probability of being associated with an entry in the address book than a caller identified by a full name;

performing the transcribing for an amount of time that is based at least in part on the probability that the caller is associated with an entry in the address book, wherein the amount of time decreases as the probability increases; and

prompting the call recipient to store at least a portion of the contact information to the recipient calling device, the prompting based at least in part on the probability that the caller is associated with an entry in the address book being below a threshold.

14. The computer program product of claim 13, wherein the stream of voice data is transcribed during the telephone conversation between the caller and the call recipient.

15. The computer program product of claim 13, wherein transcribing further comprises applying speech recognition technology.

16. The computer program product of claim 13, wherein the storing is in the address book.

19. The computer program product of claim 16, wherein the storing is during the telephone conversation.