Call context metadata

• US 9,860,355 B2
• Filed: 11/23/2015
• Issued: 01/02/2018
• Est. Priority Date: 11/23/2015
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Gorthi, Ramakrishna J., Joshi, Chandrajit G., Shah, Romil J.
• Primary Examiner(s): JACKSON, JAKIEDA R
• Application Number:
  US14/948,416
• Publication Number:
  US 20170149947A1
• Time in Patent Office: 771 Days
• Field of Search: 7042701
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 17/00 Speaker identification or verification techniques
• CPC Class Codes: G10L 17/02 Preprocessing operations, e.g. segment selection; Pattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal components; Feature selection or extraction
• CPC Class Codes: G10L 17/06 Decision making techniques; Pattern matching strategies
• CPC Class Codes: G10L 17/16 Hidden Markov models [HMM]
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces
• CPC Class Codes: G10L 25/51 for comparison or discrimination
• CPC Class Codes: H04M 1/27453 Directories allowing storage of additional subscriber data, e.g. metadata
• CPC Class Codes: H04M 1/573 Line monitoring circuits for detecting caller identification
• CPC Class Codes: H04M 1/645 with speech synthesis
• CPC Class Codes: H04M 1/656 for recording conversations
• CPC Class Codes: H04M 1/724 User interfaces specially adapted for cordless or mobile telephones
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 2201/41 using speaker recognition speaker recognition per se G10L17/00
• CPC Class Codes: H04M 2250/60 logging of communication history, e.g. outgoing or incoming calls, missed calls, messages or URLs logging of telephone numbers H04M1/56, H04M1/57
• CPC Class Codes: H04M 2250/68 with means for recording information, e.g. telephone number during a conversation recording of telephone conversations H04M1/656
• CPC Class Codes: H04M 2250/74 with voice recognition means voice activated dialling H04M1/271; voice recognition algorithms G10L15/00
• CPC Class Codes: H04M 3/42025 Calling or Called party identification service
• CPC Class Codes: H04M 3/42059 Making use of the calling party identifier
• CPC Class Codes: H04M 3/42102 Making use of the called party identifier
• CPC Class Codes: H04M 3/42221 : Conversation recording systems at the subscriber's set H04M1/656
• CPC Class Codes: H04M 3/42357 : where the information is provided to a monitoring entity such as a potential calling party or a call processing server

Abstract

A computer detects a connected voice or video call between participants and records a brief media sample. Speech recognition is utilized to determine when the call is connected as well as to transcribe the content of the audio portion of the media sample. The recorded media sample and transcribed content is associated with the connected voice or video call such that a user may reference it at a later point. The computer additionally suggests creating or editing contact information associated with the participants of the connected voice or video call based on the transcribed content.

Claims

1. A method for generating call context metadata, the method comprising:

detecting that a call is connected, wherein the call is considered connected based on speech being detected;

recording a media sample corresponding to the connected call, wherein the media sample includes an audio recording of at least a portion of the detected speech;

identifying one or more words contained within the at least the portion of the detected speech;

determining whether at least one of a phone number corresponding to the connected call and a name extracted from the one or more words is associated with a contact detailed by a contact list of a user;

based on determining that at least one of the phone number corresponding to the connected call and the name extracted from the one or more words is associated with the contact detailed by the contact list of the user, determining whether one or more fields associated with the contact are blank;

based on determining that the one or more fields associated with the contact are blank, extracting one or more attributes corresponding to the one or more blank fields from the one or more words;

ranking the extracted one or more attributes based on a probability that an attribute of the ranked one or more extracted attributes is a correct input for the one or more blank fields;

presenting the ranked one or more extracted attributes to the user as one or more selectable options for input to the one or more blank fields;

based on receiving a user input selecting at least one extracted attribute of the ranked one or more extracted attributes, associating the selected at least one extracted attribute with the one or more blank fields and deleting the media sample and the identified one or more words;

based on determining that neither the phone number corresponding to the connected call nor the name extracted from the one or more words is associated with the contact detailed by the contact list of the user, identifying one or more potential contacts based on analyzing the identified one or more words, wherein identifying one or more potential contacts further comprises;

determining a location of the user based on analyzing Global Positioning System (GPS) information associated with a device of the user, wherein determining a location of the user based on GPS information further comprises;

receiving one or more transmissions from one or more satellites detailing one or more times of transmission (TOT) and one or more transmission locations;

determining a difference in time between the TOT from the one or more satellites and one or more times of arrival (TOA) at the device of the user;

based on the determined difference in time and the one or more transmission locations, determining the location of the user;

referencing a common name list, wherein the common name list details a set of commonly used names associated with a geographic area, and wherein the geographic area is based on the determined location of the user;

determining whether at least one of the identified one or more words matches at least one of one or more names detailed by the common name list; and

presenting an option to the user to add at least one of the one of the one or more potential contacts to the contact list of the user,

wherein one or more steps of the above method are performed using one or more computers.

2. The method of claim 1, wherein the geographic area is further based on an area code of the phone number corresponding to the connected call.

3. The method of claim 1, wherein the step of identifying one or more words contained within the at least a portion of the detected speech utilizes a Hidden Markov Model.

4. The method of claim 1, further comprising:

receiving a user input to play the audio recording.

5. A computer program product for call context metadata, the computer program product comprising:

one or more computer-readable non-transitory storage media and program instructions stored on the one or more computer-readable storage media, the program instructions comprising;

program instructions to detect that a call is connected, wherein the call is considered connected based on speech being detected;

program instructions to record a media sample corresponding to the connected call, wherein the media sample includes an audio recording of at least a portion of the detected speech;

program instructions to identify one or more words contained within the at least a portion of the detected speech;

program instructions to determine whether at least one of a phone number corresponding to the connected call and a name extracted from the one or more words is associated with a contact detailed by a contact list of a user;

based on determining that at least one of the phone number corresponding to the connected call and the name extracted from the one or more words is associated with the contact detailed by the contact list of the user, program instructions to determine whether one or more fields associated with the contact are blank;

based on determining that the one or more fields associated with the contact are blank, program instructions to extract one or more attributes corresponding to the one or more blank fields from the one or more words;

program instructions to rank the extracted one or more attributes based on a probability that an attribute of the ranked one or more extracted attributes is a correct input for the one or more blank fields;

program instructions to present the ranked one or more extracted attributes to the user as one or more selectable options for input to the one or more blank fields;

based on receiving a user input selecting at least one extracted attribute of the ranked one or more extracted attributes, program instructions to associate the selected at least one extracted attribute with the one or more blank fields and program instructions to delete the media sample and the identified one or more words;

based on determining that neither the phone number corresponding to the connected call nor the name extracted from the one or more words is associated with the contact detailed by the contact list of the user, program instructions to identify one or more potential contacts based on analyzing the identified one or more words, wherein identifying one or more potential contacts further comprises;

program instructions to determine a location of the user based on analyzing Global Positioning System (GPS) information associated with a device of the user, wherein determining a location of the user based on GPS information further comprises;

program instructions to receive one or more transmissions from one or more satellites detailing one or more times of transmission (TOT) and one or more transmission locations;

program instructions to determine a difference in time between the TOT from the one or more satellites and one or more times of arrival (TOA) at the device of the user;

based on the determined difference in time and the one or more transmission locations, program instructions to determine the location of the user;

program instructions to reference a common name list, wherein the common name list details a set of commonly used names associated with a geographic area, and wherein the geographic area is based on the determined location of the user;

program instructions to determine whether at least one of the identified one or more words matches at least one of one or more names detailed by the common name list; and

program instructions to present an option to the user to add at least one of the one of the one or more potential contacts to the contact list of the user.

6. The computer program product of claim 5, wherein the geographic area is further based on an area code of the phone number corresponding to the connected call.

7. The computer program product of claim 5, wherein the program instructions to identify one or more words contained within the at least a portion of the detected speech utilizes a Hidden Markov Model.

8. The computer program product of claim 5, further comprising:

program instructions to receive a user input to play the audio recording.

9. A computer system for call context metadata, the computer system comprising:

one or more computer processors, one or more computer-readable non-transitory storage media, and program instructions stored on one or more of the computer-readable storage media for execution by at least one of the one or more processors, the program instructions comprising;

program instructions to detect that a call is connected, wherein the call is considered connected based on speech being detected;

program instructions to record a media sample corresponding to the connected call, wherein the media sample includes an audio recording of at least a portion of the detected speech;

program instructions to identify one or more words contained within the at least a portion of the detected speech;

program instructions to determine whether at least one of a phone number corresponding to the connected call and a name extracted from the one or more words is associated with a contact detailed by a contact list of a user;

based on determining that at least one of the phone number corresponding to the connected call and the name extracted from the one or more words is associated with the contact detailed by the contact list of the user, program instructions to determine whether one or more fields associated with the contact are blank;

based on determining that the one or more fields associated with the contact are blank, program instructions to extract one or more attributes corresponding to the one or more blank fields from the one or more words;

program instructions to rank the extracted one or more attributes based on a probability that an attribute of the ranked one or more extracted attributes is a correct input for the one or more blank fields;

program instructions to present the ranked one or more extracted attributes to the user as one or more selectable options for input to the one or more blank fields;

based on receiving a user input selecting at least one extracted attribute of the ranked one or more extracted attributes, program instructions to associate the selected at least one extracted attribute with the one or more blank fields and program instructions to delete the media sample and the identified one or more words;

based on determining that neither the phone number corresponding to the connected call nor the name extracted from the one or more words is associated with the contact detailed by the contact list of the user, program instructions to identify one or more potential contacts based on analyzing the identified one or more words, wherein identifying one or more potential contacts further comprises;

program instructions to determine a location of the user based on analyzing Global Positioning System (GPS) information associated with a device of the user, wherein determining a location of the user based on GPS information further comprises;

program instructions to receive one or more transmissions from one or more satellites detailing one or more times of transmission (TOT) and one or more transmission locations;

program instructions to determine a difference in time between the TOT from the one or more satellites and one or more times of arrival (TOA) at the device of the user;

based on the determined difference in time and the one or more transmission locations, program instructions to determine the location of the user;

program instructions to reference a common name list, wherein the common name list details a set of commonly used names associated with a geographic area, and wherein the geographic area is based on the determined location of the user;

program instructions to determine whether at least one of the identified one or more words matches at least one of one or more names detailed by the common name list; and

program instructions to present an option to the user to add at least one of the one of the one or more potential contacts to the contact list of the user.

10. The computer system of claim 9, wherein the geographic area is further based on an area code of the phone number corresponding to the connected call.

11. The computer system of claim 9, wherein the program instructions to identify one or more words contained within the at least a portion of the detected speech utilizes a Hidden Markov Model.

12. The computer system of claim 9, further comprising:

program instructions to receive a user input to play the audio recording.