A training method and device for an automated assistant
A training method and assistant technology, applied in the field of information processing, can solve the problems of limited functions and training methods, poor user physical examination, etc., and achieve the effect of improving user experience, optimizing automatic learning ability, and optimizing automatic processing of user tasks.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0034] refer to figure 1 , Embodiment 1 of the present invention provides a training method for an automated assistant, characterized in that:
[0035] Step 101, collecting calling phone number information at the time of incoming calls, classifying and storing the calling numbers, and setting a calling number attribute matching relationship table for the calling numbers;
[0036] Step 102, collecting call voice information, and classifying and storing the voice information according to the calling phone number;
[0037] Step 103, identifying the voice information, distinguishing the dialogue subject, and marking the dialogue data in the voice information according to the telephone number corresponding to the dialogue subject;
[0038] Wherein step 103 further includes, storing the dialogue data in the text form after marking in the call-in data resource library, wherein the caller's phone number is used as the primary index, and the call end time of the call-in phone is taken...
Embodiment 2
[0095] refer to image 3, on the basis of Embodiment 1, Embodiment 2 provides an intelligent response method based on identity confirmation, wherein the method includes but is not limited to:
[0096] Step 301, receiving call information, and determining the calling phone number according to the call information;
[0097] Step 302: Obtain attribute information of the calling phone number according to the calling phone number;
[0098] Step 303, judging whether the attribute information of the calling phone number belongs to the first attribute category; if the attribute information of the calling phone number belongs to the first attribute category, perform step 305, if the attribute information of the calling phone number does not belongs to the first attribute category, go to step 304;
[0099] Step 304, start the first timer, if the first timer expires and the user does not answer, execute step 305;
[0100] Step 305, perform call takeover, and determine the logic of the...
Embodiment 3
[0125] refer to Figure 5 , on the basis of the foregoing embodiments, Embodiment 3 of the present invention also provides an automatic learning method for a voice assistant;
[0126] Step 501, collecting the information of the called phone number when making an outgoing call, obtaining the called number of the outgoing call, and determining the number attribute of the called number;
[0127] Step 502, storing the call voice information after the call is connected, and classifying and storing the call voice information according to the called phone number;
[0128] Step 503, identifying the voice information, distinguishing the dialogue subject, and marking the dialogue data in the voice information according to the telephone number corresponding to the dialogue subject;
[0129] Step 504, query the application log to determine whether there is an application operation record and / or application notification record within the first predetermined time range before the call star...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More