VOICEPRINT RECOGNITION METHOD AND DEVICE, TERMINAL DEVICE AND STORAGE MEDIUM

摘要

Provided are a voiceprint recognition method and device, a terminal device and a storage medium, the method comprising: allocating each agent to a corresponding group according call permissions (S1); collecting and processing the voiceprint features of each agent in the group to obtain an integrated voiceprint feature of the group (S2); constructing a dynamic voiceprint recognition model on the basis of the groups, the call permissions and the integrated voiceprint feature (S3); when a call operation from an agent is detected, obtaining voice data of the agent and extracting voiceprint features (S4); on the basis of the dynamic voiceprint recognition model, determining the call permissions of the agent (S5); and if the call permissions of an agent to be recognized does not meet a customer level requirement, taking preset early-warning processing measures (S6), thereby carrying out monitoring. The group in which the agent is located is determined by means of voiceprint recognition, and real-time monitoring is realized according to the correspondence between the group and the permissions, thereby effectively improving the real-time performance of monitoring and the call quality of the agent call.