Streaming voice response method, device and voice calling robot

Abstract

The present invention provides a method and device for responding to a streaming voice and a voice calling robot thereof, including: acquiring a streaming voice signal based on a target call, and determining key data features; matching the key data features in a recognition mapping relationship to obtain The reason for the call failure related to the key data feature; match the cause of the call failure in the response mapping relationship to perform the relevant response operation; the key data feature is the voice feedback prompt information jumped when the call is not connected; the response operation at least includes Disconnect the call. The present invention obtains the streaming voice signal through the call with the called party, converts the streaming voice signal into text information, determines key data features, obtains the reason for the failure of the call, and implements the execution of the relevant response operation based on the reason for the failure of the call. The present invention can realize the judgment of the current outgoing call state quickly, and cut off the task of call failure in time, thereby releasing the line resources and improving the utilization rate of the agent line.

Description

technical field [0001] The present invention relates to the field of speech recognition response, in particular to a method and device for responding to streaming speech and a voice calling robot thereof. Background technique [0002] At present, with the development of artificial intelligence technology, the traditional method of outbound calls through manual agents consumes a lot of labor costs, which is no longer suitable for the development of today's society. In the process of intelligent outbound calls, there is a high probability of failure of the call, and the overall connection rate is not high, usually due to the status of the called party's "empty number", "off", "calling" and other states. lead to. Based on this situation, the intelligent outbound call robot cannot quickly judge the current state of outbound calls, and cannot cut off the tasks that fail to call, and thus cannot release line resources, thereby reducing the utilization rate of agent lines. [000...

AI Technical Summary

Problems solved by technology

[0002] At present, with the development of artificial intelligence technology, the traditional way of making outbound calls through manual agents consumes a lot of labor costs, which is no longer suitable for the development of today's society. Instead, intelligent outbound robots, which can automate and mass-produce outbound calls, are used.

In the process of intelligent outbound calls, there is a high probability of encountering call failures, and the overall connection rate is not high, usually due to the status of the called party such as "empty number", "off phone" and "in progress". lead to

Based on this situation, the intelligent outbound call robot cannot quickly judge the current outbound call status, cannot cut off the failed call task, and cannot release line resources, thereby reducing the utilization rate of the agent line

[0003] For robot intelligent outbound call scenarios, the current mainstream technical means are to judge the cause of phone call failure through communication signaling or neural network models. However, these two methods have certain limitations. In terms of methods, on the one hand, the way to obtain signaling is relatively difficult; on the other hand, due to the complicated relationship between the switch equipment of the operator, even if it can be obtained, the obtained results are not very accurate, and the determined telephone call failure The reason is also more abstract

For the method of neural network model, there is a problem that the output result is relatively fixed and not flexible enough. Once the model is trained, it cannot be adjusted or it takes a lot of time to relabel the data training model.

[0004] At present, there is no effective method for judging the cause of the current call failure, thereby releasing line resources and improving line utilization, and there is no technical solution that can solve the above technical problems. Specifically, there is no flow Voice response method, device and voice call robot thereof

Images