Speech emotion-based verbal skill pushing method and device, equipment and storage medium

Abstract

The invention discloses a speech emotion-based verbal skill pushing method and device, equipment and a storage medium, and the method comprises the steps: detecting a voice replied by a user, and obtaining a target voice; recognizing the target voice, and obtaining a target emotion contained in the target voice; obtaining a target value of the target emotion based on the algorithm model; and comparing the target value with a threshold value, and determining a pushed target verbal skill. According to the technical scheme, after it is judged that the first-time persuasion talk is failed, the voice replied by the user can be collected and subjected to emotion recognition processing, then the negative emotions are classified and refined, the corresponding weights are correspondingly given to all the negative emotions through the corpus retrieval algorithm, and then the negative emotions of the user can be obtained through the algorithm. And calculating whether the threshold value of the corresponding persuasion verbal skill is reached or not, and recommending the corresponding persuasion verbal skill if the threshold value of the corresponding persuasion verbal skill is reached, so that the combination of the anti-fraud system and the recommendation algorithm is realized, and the corresponding persuasion verbal skill is correspondingly recommended for the emotion after the voice of the user is refined.

Description

technical field [0001] The invention belongs to the technical field of computers, and in particular relates to a method, a device, a device and a storage medium for pushing words based on voice emotions. Background technique [0002] In the anti-fraud system, the existing solution is to use voice emotion recognition technology to roughly divide the user's emotions into three categories: negative, neutral, and positive; And push a unified dissuading speech technique. At this time, emotion recognition will be performed on the user's reaction words, and when the user's voice emotion is recognized as negative, the fraud warning will be directly pushed to the relatives of the user to prevent them from being deceived; specific , when the system recognizes that the call may be a fraudulent call and pushes a unified dissuading message, once the user's voice emotion recognition result is negative, it will judge that the dissuasion has failed, and push a warning of possible fraud to t...

AI Technical Summary

Problems solved by technology

[0003] However, there is no corresponding second dissuading speech based on the user's reaction to the dissuading speech, and the fraud warning is directly sent to the user's relatives. This cannot be targeted and suitable for the user's current mood. If the dissuasion skills are too good, it will not achieve a better dissuasion effect; and if a dissuasion fails and a warning is sent, the possibility of making a wrong judgment will also increase

Images