Computing device for providing dialogues services
AssigneeActionPower Corp.
InventorsJihwa Lee, Hongsik Jo
Life of this patent
Estimating the term…
Abstract
Disclosed is a computer program executable by one or more processors and stored in a computer-readable storage medium, the computer program causing the one or more processors to perform one or more operations below, the operations including: an operation of receiving first voice data from a first user terminal and receiving second voice data from a second user terminal; an operation of generating integrated voice data based on the first voice data and the second voice data; and an operation of generating the integrated voice data as a conversation record by using a voice recognition model.
Claims17
A non-transitory computer-readable storage medium storing a computer program executable by one or more processors, the computer program causing the one or more processors to perform one or more operations below, the operations comprising:
- an operation of receiving first voice data from a first user terminal and receiving second voice data from a second user terminal;
- an operation of generating integrated voice data by using a method of dividing a speaker based on a speaker utterance time ratio of the first voice data and the second voice data;
- and
- an operation of generating the integrated voice data as a conversation record by using a voice recognition model,
- wherein the method of dividing the speaker based on the speaker utterance time ratio of the first voice data and the second voice data includes at least one of:
- an operation of calculating an utterance time ratio for each speaker based on a speaker of a frame excluding an overlapping section to divide the speakers for the overlapping section when the overlapping section is identified;
- and
- an operation of dividing the speaker of the overlapping section based on an utterance time ratio for each speaker for a previous frame of an overlapping section frame when the overlapping section is identified, and
- wherein the operation of dividing the speaker of the overlapping section based on an utterance time ratio for each speaker for a previous frame of an overlapping section frame when the overlapping section is identified is performed when the precious frame of the overlapping section frame is equal to or greater than a selected ratio of voice data.
A method for providing a conversation record service, the method comprising:
- receiving first voice data from a first user terminal and receiving second voice data from a second user terminal;
- generating integrated voice data by using a method of dividing a speaker based on a speaker utterance time ratio of the first voice data and the second voice data;
- and
- generating the integrated voice data as a conversation record by using a voice recognition model,
- wherein the method of dividing the speaker based on the speaker utterance time ratio of the first voice data and the second voice data includes at least one of:
- calculating an utterance time ratio for each speaker based on a speaker of a frame excluding an overlapping section to divide the speakers for the overlapping section when the overlapping section is identified;
- and
- dividing the speaker of the overlapping section based on an utterance time ratio for each speaker for a previous frame an overlapping section frame when the overlapping section is identified, and
- wherein the dividing the speaker of the overlapping section based on an utterance time ratio for each speaker for a previous frame of an overlapping section frame when the overlapping section is identified is performed when the previous frame of the overlapping section frame is equal to or greater than a selected ratio of voice data.
A server for providing a conversation record service, the server comprising:
- a server processor including one or more cores;
- a server memory for storing program codes executable in the server processor;
- and
- a server network unit for transceiving data with one or more user terminals,
- wherein the processor is configured to:
- receive first voice data from a first user terminal and receiving second voice data from a second user terminal,
- generate integrated voice data by using a method of dividing a speaker based on a speaker utterance time ratio of the first voice data and the second voice data, and
- generate the integrated voice data as a conversation record by using a voice recognition model,
- wherein the method of dividing the speaker based on the speaker utterance time ratio of the first voice data and the second voice data includes at least one of:
- calculating an utterance time ratio for each speaker based on a speaker of a frame excluding an overlapping section to divide the speakers for the overlapping section when the overlapping section is identified;
- and
- dividing the speaker of the overlapping section based on an utterance time ratio for each speaker for a previous frame of an overlapping section frame when the overlapping section is identified, and
- wherein the dividing the speaker of the overlapping section based on an utterance time ratio for each speaker for a previous frame of an overlapping section frame when the overlapping section is identified is performed when the previous frame of the overlapping section frame is equal to or greater than a selected ratio of voice data.
Dependencies are inferred from the claim text.