Method and apparatus for processing voice information to determine emotion based on volume and pacing of the voice
AssigneeAlibaba Group Holding Limited
InventorQing Ling
Life of this patent
Estimating the term…
Abstract
A method and an apparatus for processing voice information are provided by the present application. The method comprises: obtaining the voice data of a user; performing an emotion analysis on the voice data which comprises multiplying volume and pacing of the voice data to help determine the priority of the voice data; and connecting the user to a corresponding service channel according to the priority of the voice data. The present application can improve the service quality of IVR.
Claims18
A method for processing voice information, comprising:
- obtaining voice data of a user in an intelligent channel configured to play an automated voice prompt;
- performing text conversion on the voice data to determine a category of problem from categories including invalid user, inquiry user, service requesting user and urgent request user;
- performing an emotion analysis on the voice data to determine an emotion value, wherein performing the emotion analysis on the voice data comprises performing numerical processing on volume and pacing of the voice data, wherein the performing numerical processing on volume and pacing of the voice data comprises multiplying the volume and the pacing;
- determining that the emotion value reaches a threshold;
- and
- connecting, in response to the determining that the emotion value reaches a threshold and determining that the category of problem is a service requesting user or an urgent request user, the user to a manual channel attended by a human.
A non-transitory computer readable medium comprising program instructions which, when executed by a computer, cause the computer to perform a method comprising:
- obtaining voice data of a user in an intelligent channel configured to play an automated voice prompt;
- performing text conversion on the voice data to determine a category of problem from categories including invalid user, inquiry user, service requesting user and urgent request user;
- performing an emotion analysis on the voice data to determine an emotion value, wherein performing the emotion analysis on the voice data comprises performing numerical processing on volume and pacing of the voice data, wherein the performing numerical processing on volume and pacing of the voice data comprises multiplying the volume and the pacing;
- determining that the emotion value reaches a threshold;
- and
- connecting, in response to the determining that the emotion value reaches a threshold and determining that the category of problem is a service requesting user or an urgent request user, the user to a manual channel attended by a human.
A method for processing voice information, comprising:
- obtaining voice data of a user;
- performing an emotion analysis on the voice data to determine an emotion value, wherein performing the emotion analysis on the voice data comprises performing numerical processing on volume and pacing of the voice data, and the emotional value is determined at least based on multiplying the volume and the pacing;
- determining a priority of the voice data according to the emotional value based on a preset mapping relationship between the emotion value and the priority;
- and
- connecting the user to a corresponding service channel according to the priority of the voice data.
Dependencies are inferred from the claim text.