This paper presents a semi-structured interview system composed of multiple LLM components, in which an LLM-based interviewer conducts dialogue and another LLM estimates interviewee persona attributes from the dialogue history and dynamically updated slot information. The system prioritizes questions related to important yet unconfirmed attributes to support efficient interviews. We evaluate persona attribute estimation performance when interviewing both human participants and an LLM-based pseudo interviewee under zero-shot and few-shot prompting conditions. Experiments across three domains (IT engineer, store staff, and school teacher) with a total of 39 interviews show that persona attribute estimation is more challenging for human interviewees than for the LLM pseudo interviewee under zero-shot prompting. However, introducing few-shot prompting consistently improves estimation performance for human interviewees, primarily by increasing recall. On the other hand, its effect on the LLM pseudo interviewee is mixed across domains. A case study confirms, through case-based analysis, that persona attribute estimation improves when illustrative examples are provided.