Manuscript received September 20, 2025; accepted November 7, 2025; published December 22, 2025.
Abstract—Aiming to address issues in the transcription of police interrogations such as inefficiency, lack of standardization, and poor information coordination, this paper designs an intelligent interrogation transcription system based on a large language model. By constructing a hierarchically decoupled system architecture that integrates the data layer, model layer, service layer, and application layer, the system achieves intelligent processing of the entire interrogation workflow. It innovatively implements core functions such as adaptive transcription generation and dynamic completion for cases, multimodal interaction and active interrogation assistance, compliance review, and analysis of evidential contradictions. The system ensures professional reliability through fine-tuning of judicial large language models and secure deployment techniques. Experimental results demonstrate that the system can reduce transcription time by over 50% while significantly improving transcription completeness and procedural compliance rates. This provides a feasible technical solution for the “technology-driven policing” initiative and the standardization of law enforcement in public security organs. Furthermore, the “human-machine collaborative” smart case handling model offers a replicable practical pathway for the digital transformation of policing operations.
Keywords—intelligent interrogation transcription system for public security, large language model, judicial domain fine-tuning, human-machine collaboration
Cite: Junquan Zhou, Shanting Song, Shuifeng Zhang, Qingyang Gao, Yuekai Ma, Zhengnan Tian, and Buyun Chen, "Research and Exploration of an Intelligent Interrogation Transcription System for Public Security Based on Large Language Models," International Journal of Engineering and Technology, vol. 17, no. 4, pp. 221-225, 2025.
Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (
CC BY 4.0).
