Classifying spoken content in a teleconference
View Patent ↗A method and an apparatus for classifying spoken content in a teleconference for a follower of the teleconference is disclosed. The method comprises: detecting a topic to which the spoken content belongs; determining a (overall) correlation degree between the follower and the spoken content at least according to a correlation degree between the follower and the topic; and classifying the spoken content according to the (overall) correlation degree between the follower and the spoken content. With the method and the apparatus, the correlation degree between the spoken content in the teleconference and the follower of the teleconference can be determined automatically, and the spoken content can be classified according to the correlation degree, so that the follower can selectively pay attention to some spoken contents during the teleconference, which reduces a burden of the follower and improves conference efficiency.
1. A method for classifying spoken content in a teleconference for a follower of the teleconference, the method comprising:
detecting, via voice recognition and semantic analysis executed by one or more processors, a topic to which the spoken content belongs;
determining, by the one or more processors, a correlation degree between the follower and the topic to provide a topic correlation degree;
determining, by the one or more processors, a correlation degree between the topic and the spoken content to provide a speaker correlation degree;
classifying, by the one or more processors, the spoken content into a plurality of correlation degree levels according to an overall correlation degree that is a weighted sum of the topic correlation degree and the speaker correlation degree divided by an elapsed time since the spoken content was spoken; and
displaying the spoken content according to the plurality of correlation degree levels.
2. The method of claim 1 , wherein the topic correlation degree is calculated prior to the teleconference according to properties of the follower and properties of the topic.
3. The method of claim 1 , wherein the topic correlation degree is determined by fuzzy matching on properties of the follower and properties of the topic.
4. The method of claim 1 , wherein the topic correlation degree is determined by determining if each of a plurality of properties of the topic matches one or more properties of the follower and adding a property score to the topic correlation degree in response to a match of a property.
5. The method of claim 1 , wherein the correlation degree determining module determines the topic correlation degree by determining if each of a plurality of properties of the topic matches one or more properties of the follower and adding a property score to the topic correlation degree in response to a match of a property.
6. A computer program product for classifying spoken content in a teleconference for a follower of the teleconference, the computer program product comprising:
at least one non-transitory computer readable storage medium, and program instructions stored on the at least one non-transitory computer readable storage medium, the program instructions comprising instructions to:
detect, via voice recognition and semantic analysis executed by one or more processors, a topic to which the spoken content belongs;
determine, by the one or more processors, a correlation degree between the follower and the topic to provide a topic correlation degree;
determine, by the one or more processors, a correlation degree between the topic and the spoken content to provide a speaker correlation degree;
classify, by the one or more processors, the spoken content into a plurality of correlation degree levels according to an overall correlation degree that is a weighted sum of the topic correlation degree and the speaker correlation degree divided by an elapsed time since the spoken content was spoken; and
display the spoken content according to the plurality of correlation degree levels.
7. The computer program product of claim 6 , wherein the topic correlation degree is calculated prior to the teleconference according to properties of the follower and properties of the topic.
8. The computer program product of claim 6 , wherein the program instructions comprising instructions to determine the topic correlation degree by fuzzy matching on properties of the follower and properties of the topic.
9. The computer program product of claim 6 , wherein the program instructions comprising instructions to determine the topic correlation degree by determining if each of a plurality of properties of the topic matches one or more properties of the follower and adding a property score to the topic correlation degree in response to a match of a property.
10. An apparatus for classifying spoken content in a teleconference for a follower of the teleconference, comprising:
one or more processors;
a plurality of program modules executed by the one or more processors including a detecting module, a correlation degree determining module, and a classifying module;
the detecting module configured to detect, via voice recognition and semantic analysis, a topic to which the spoken content belongs;
the correlation degree determining module configured to determine a correlation degree between the follower and the topic to provide a topic correlation degree;
the correlation degree determining module further configured to determine a correlation degree between the topic and the spoken content to provide a speaker correlation degree;
the classifying module configured to classify the spoken content into a plurality of correlation degree levels according to an overall correlation degree that is a weighted sum of the topic correlation degree and the speaker correlation degree divided by an elapsed time since the spoken content was spoken; and
the classifying module further configured to display the spoken content according to the plurality of correlation degree levels.
11. The apparatus of claim 10 , wherein the topic correlation degree is calculated prior to the teleconference according to properties of the follower and properties of the topic.
12. The apparatus of claim 10 , wherein the correlation degree determining module determines the topic correlation degree by fuzzy matching on properties of the follower and properties of the topic.