Search

CN-121996930-A - Real-time communication auxiliary method and system based on digital personality model

CN121996930ACN 121996930 ACN121996930 ACN 121996930ACN-121996930-A

Abstract

The application discloses a real-time communication auxiliary method and a system based on a digital personality model, which are used for receiving scene setting input by a user and acquiring multi-mode data in a communication scene in real time; the method comprises the steps of preprocessing multi-modal data, extracting multi-modal personality characteristics from the preprocessed multi-modal data, aggregating the multi-modal personality characteristics, inputting the multi-modal personality characteristics into a pre-built digital personality model for preliminary personality assessment to obtain a personality trait prediction result of a user, and matching and generating a real-time communication strategy from a pre-built strategy database according to scene setting and the personality trait prediction result. The method provided by the application can be used for real-time insight into the psychological state and personality traits of the opposite party in the communication process, and generating and transmitting the optimal communication strategy in real time so as to reduce potential conflict, and solve the problems of single function, serious hysteresis and lack of individuality of the traditional communication auxiliary method.

Inventors

  • QIAN YU
  • LI YICHENG
  • SUN ZHONGKAI

Assignees

  • 北京齿伦转动科技有限公司

Dates

Publication Date
20260508
Application Date
20260126

Claims (10)

  1. 1. The real-time communication auxiliary method based on the digital personality model is characterized by comprising the following steps of: step 1, receiving scene setting input by a user, and acquiring multi-mode data in a communication scene in real time, wherein the multi-mode data comprises video data, audio data and text data; Step 2, carrying out data preprocessing on the multi-modal data, and extracting multi-modal personality characteristics from the preprocessed multi-modal data; Step 3, aggregating the multi-modal personality characteristics, inputting the multi-modal personality characteristics into a pre-built digital personality model for preliminary personality assessment to obtain a personality trait predictive result of the user; And 4, matching and generating a real-time communication strategy from a pre-constructed strategy database according to the scene setting and the personality trait prediction result.
  2. 2. The method for assisting real-time communication based on a digital personality model according to claim 1, further comprising real-time analyzing a degree of matching between a behavior pattern of an opposite party and the digital personality model, triggering collision early warning in real time when emotion fluctuation, a collision limb language or a collision language is detected, and adjusting a generated real-time communication strategy in real time according to the collision early warning.
  3. 3. The method for assisting real-time communication based on a digital personality model according to claim 1, further comprising anonymizing communication data with a user of the established profile each time, and storing the anonymized communication data in a corresponding character file continuously to obtain a long-term history file of the user of the established profile.
  4. 4. The method for assisting real-time communication based on a digital personality model according to claim 3, further comprising receiving a communication target input by a user, retrieving a long-term history file of the user according to the communication target, and performing a emotion Jing Yuyan, and generating a communication gist and risk prompt based on long-term observation according to the scenario previewing result.
  5. 5. The method for assisting real-time communication based on a digital personality model of claim 3, further comprising automatically calculating a relationship health score for each persona based on the user's long-term history file and comprehensively considering the relationship network for displaying the user via the visual map.
  6. 6. The method according to claim 1, wherein in step 1, the scenario setting includes personality tags of communication objects, my characters and appeal bottom lines, conversation topics and targets, and scene information.
  7. 7. The method according to claim 1, wherein in step 2, the computer vision technique, the voice signal processing technique and the NLP technique are used for extracting the multi-modal personality characteristics from the preprocessed multi-modal data.
  8. 8. The method for assisting real-time communication based on a digital personality model according to claim 1, wherein in step 4, the policy database is constructed by using a knowledge graph.
  9. 9. The method according to claim 1, wherein in step 4, the real-time communication strategy includes speaking, topics to be avoided, recommended speech utterances, and recommended limb language coordination.
  10. 10. A digital personality model-based real-time communication assistance system, comprising: the system comprises a multi-mode data acquisition module, a communication module and a communication module, wherein the multi-mode data acquisition module is used for receiving scene setting input by a user and acquiring multi-mode data in a communication scene in real time, and the multi-mode data comprises video data, audio data and text data; the multi-modal personality characteristic extraction module is used for carrying out data preprocessing on the multi-modal data and extracting multi-modal personality characteristics from the preprocessed multi-modal data; The personality recognition module is used for aggregating the multi-modal personality characteristics, inputting the multi-modal personality characteristics into a pre-built digital personality model for preliminary personality assessment to obtain a personality trait prediction result of the user; And the communication strategy generation module is used for matching and generating a real-time communication strategy from a pre-constructed strategy database according to the scene setting and the personality trait prediction result.

Description

Real-time communication auxiliary method and system based on digital personality model Technical Field The application relates to the technical field of man-machine interaction, in particular to a real-time communication auxiliary method and system based on a digital personality model. Background With the rapid development of artificial intelligence, natural Language Processing (NLP), emotion computing, and other technologies, communication assistance technologies are widely used in various fields such as customer service, sales, psychological consultation, and enterprise management. The prior art focuses mainly on identifying the emotional state of a user by means of single-mode data such as text, voice or video and on the basis of which basic feedback is provided. However, the current mainstream communication assistance technology still has significant drawbacks, and it is difficult to meet the deep support requirements in a real complex interpersonal interaction scenario, which is specifically expressed in the following four aspects: 1. The function is single, most of the existing communication auxiliary tools only have emotion recognition capability, and active guidance on communication strategies is lacking. Even if the emotion of the opposite party can be accurately judged to be 'dissatisfied' or 'hesitant', a specific coping proposal or speaking optimization scheme cannot be generated, so that a user still needs to rely on self experience to make a decision. 2. The response is seriously lagged, that is, most of emotion analysis and communication evaluation methods at present adopt an offline processing mode, and a multi-disc report can be generated only after the dialogue is ended. The analysis of 'post Zhu Geliang' can not provide real-time intervention on key nodes (such as about to run away, negotiating and falling into the dead office, etc.) of communication, and the best opportunity for influencing the trend of communication is missed, so that the actual utility of the technology in dynamic interaction is weakened. 3. The lack of personalized adaptation capability-general chat robots or auxiliary plugins are typically based on popular corpus training and are not modeled for personality traits, historical interaction preferences, cultural backgrounds, or current contexts of particular communication objects. Therefore, the advice provided by the method is often on the surface and templated, and even communication friction can be aggravated by the "one-tool" technique, so that the real "person-induced" cannot be realized. 4. The growth and relation evolution insights are lacking, and the existing CRM or communication recording system can store a large amount of interaction logs, but generally lacks the capability of fusion analysis of micro emotion signals of multiple modes (such as voice intonation, facial expression, text semantics and behavior tracks), and does not have a long-term learning mechanism to track the change trend of implicit indexes such as relation affinity, trust and the like. The system can not refine deep rules from historical interaction, can not dynamically adjust communication strategies along with relation development, and is difficult to support long-term and high-value interpersonal relation management. In summary, the current communication assistance technology has obvious shortages in terms of functionality, timeliness, individuation, continuous evolution capability, and the like, and a new generation intelligent communication assistance method capable of integrating multi-modal sensing, real-time intervention, personality modeling and long-term learning capability is needed to truly realize closed-loop communication support of understanding emotion, knowing objects, giving advice and growing. Disclosure of Invention Therefore, the application provides a real-time communication auxiliary method and a system based on a digital personality model, which are used for solving the problems of single function, serious hysteresis and lack of individuality of the communication auxiliary method in the prior art. In order to achieve the above object, the present application provides the following technical solutions: in a first aspect, a method for assisting real-time communication based on a digital personality model includes: step 1, receiving scene setting input by a user, and acquiring multi-mode data in a communication scene in real time, wherein the multi-mode data comprises video data, audio data and text data; Step 2, carrying out data preprocessing on the multi-modal data, and extracting multi-modal personality characteristics from the preprocessed multi-modal data; Step 3, aggregating the multi-modal personality characteristics, inputting the multi-modal personality characteristics into a pre-built digital personality model for preliminary personality assessment to obtain a personality trait predictive result of the user; And 4, matching and generating a real-time communi