Knowledge sharing based on meeting information

Patent metadata

Field	Value
Publication number	US-9734208-B1
Application number	US-201313893215-A
Country	US
Kind code	B1
Filing date	May 13, 2013
Priority date	May 13, 2013
Publication date	Aug 15, 2017
Grant date	Aug 15, 2017

Abstract

Features are disclosed for automatically facilitating knowledge sharing using information collected during meetings. Collected information may include both the content and context of a meeting. The meeting content may comprise text collected by automatic speech recognition or entered manually at a user device. The meeting context may comprise information such as the identities of meeting participants and subject matter from a meeting agenda. The content and context information may be used to identify relevant topics for which knowledge is available, and the available knowledge may be presented to one or more of the meeting participants.

First claim

What is claimed is:

1. A system comprising:
a first physical data store that stores context information associated with a meeting of meeting participants;
a second physical data store that stores an index of topics for the meeting; and a server in communication with the first physical data store and the second physical data store, the server including one or more hardware processors that execute computer-executable instructions to:
receive, from a user device, captured audio data corresponding to speech of a spoken dialogue involving two or more meeting participants;
perform automatic speech recognition to generate a transcript of the speech of the spoken dialogue involving the two or more meeting participants from the captured audio data, the transcript comprising a plurality of words;
compare at least some of the plurality of words with keywords in the index of topics stored in the second physical data store to identify candidate topics from the index;
filter, using context information associated with the meeting and stored in the first physical data store, the candidate topics to identify a first topic and a second topic relevant to the meeting from the candidate topics;
generate, from the transcript, an annotated transcript with a first embedded indication of the first topic and a second embedded indication of the second topic, wherein the first embedded indication is configured for presentation as a textual representation of the first topic, the textual representation different from a visual representation of another portion of the annotated transcript;
send, to the user device, the annotated transcript;
receive, from the user device, input data representing a user selection to cause presentation of additional information associated with the first topic;
generate knowledge results related to the first topic, wherein the knowledge results are different than the annotated transcript; and
send, to the user device, the knowledge results.

2. The system of claim 1 , the context information comprising at least one of identification of a particular meeting participant, meeting agenda, meeting subject matter, meeting date, meeting time, or meeting location.

3. The system of claim 1 , wherein the one or more hardware processors further execute the computer-executable instructions to generate the knowledge results by:
obtaining excerpts of information related to the first topic from a plurality of information sources; and
organizing the excerpts into at least one of a summary, digest or list.

4. The system of claim 1 , wherein the first physical data store and the second physical data store are portions of the same physical data store.

5. A computer-implemented method comprising:
under control of one or more computing devices configured with specific computer-executable instructions,
receiving an electronic transcript of a conversation of spoken dialogue involving two or more participants, the electronic transcript comprising a plurality of words;
determining an attribute of the electronic transcript from at least some of the plurality of words;
searching, using the attribute, a collection of topics stored in an electronic data store to identify a first topic and a second topic relevant to the conversation;
generating, from the electronic transcript, an annotated electronic transcript comprising a first embedded indication of the first topic and a second embedded indication the second topic, wherein the first embedded indication is configured for presentation as a textual representation of the first topic to indicate availability of electronically stored data about the first topic, the textual representation different from a visual representation of another portion of the annotated electronic transcript;
sending the annotated electronic transcript to a computing device associated with a participant of the conversation;
receiving, from the computing device, input data representing a user selection to cause presentation of additional information associated with the first topic; and
sending a portion of the electronically stored data about the first topic, to the computing device, wherein the portion of the electronically stored data is different from the annotated electronic transcript and represents a portion of shared knowledge.

6. The computer-implemented method of claim 5 , the electronic transcript being generated at least in part by automatic speech recognition.

7. The computer-implemented method of claim 5 , the electronic transcript comprising text input received from the computing device associated with the participant of the conversation.

8. The computer-implemented method of claim 5 , the attribute comprising context information associated with the conversation.

9. The computer-implemented method of claim 8 , the context information comprising at least one of identification of a meeting participant, meeting agenda, meeting subject matter, meeting date, meeting time, or meeting location.

10. The computer-implemented method of claim 5 , wherein sending the portion of the electronically stored data about the first topic comprises:
retrieving the electronically stored data from the electronic data store;
generating a summary of the electronically stored data, wherein the summary is different than the annotated electronic transcript; and
sending the summary of the electronically stored data to the computing device associated with the participant of the conversation.

11. The computer-implemented method of claim 5 , wherein identifying at least the first topic further comprises:
retrieving context information associated with the conversation; and
determining that the participant of the conversation is likely interested in the first topic based at least in part on the context information associated with the conversation.

12. The computer-implemented method of claim 5 , wherein identifying at least the first topic further comprises:
estimating a level of relevance of the first topic for the participant; and
determining that the level of relevance satisfies a threshold.

13. The computer-implemented method of claim 5 , wherein determining the attribute of the electronic transcript comprises looking up an attribute corresponding to a word of the electronic transcript.

14. A non-transitory computer-readable medium comprising instructions executable by one or more processors to at least:
receive at least a portion of a transcript from a conversation of spoken dialogue involving two or more participants;
receive metadata associated with the conversation, the metadata comprising at least one topic or keyword;
search, using the metadata, a collection of topics stored in an electronic data store to identify a first topic and a second topic relevant to the conversation;
generate an annotated portion of the transcript comprising a first embedded indication of the first topic and a second embedded indication of the second topic, wherein the first embedded indication is configured for presentation as a textual representation of the first topic to indicate availability of electronically stored data about the first topic, the textual representation different from a visual representation of another portion of the transcript;
send, while the conversation is in process, the annotated portion of the transcript to a computing device associated with a participant of the conversation;
receive, from the computing device, input data representing a user selection to cause presentation of additional information associated with the first topic; and
send, while the conversation is in process, a portion of the electronically st

Assignees

• Audible Inc

Inventors

• Goldstein Douglas Scott
• Arora Ajay
• Brothers Nathan Garret
• Hwang Douglas Cho
• Story Jr Guy Ashley

Classifications

• G06Q10/10 Primary

  Office automation; Time management · CPC title

• G06F17/3053 Primary

  Physics · mapped topic

• G10L15/26

  Speech to text systems (G10L15/08 takes precedence) · CPC title

• G06F16/24578 Primary

  using ranking · CPC title

• G06Q10/101

  Collaborative creation, e.g. joint development of products or services · CPC title