Saved in:
Bibliographic Details
Main Authors: Yang, Xingjian, Yang, Yudong, Guo, Zhixing, Zhou, Yongjie, Yan, Nan, Wang, Lan
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.10161
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918439592919040
author Yang, Xingjian
Yang, Yudong
Guo, Zhixing
Zhou, Yongjie
Yan, Nan
Wang, Lan
author_facet Yang, Xingjian
Yang, Yudong
Guo, Zhixing
Zhou, Yongjie
Yan, Nan
Wang, Lan
contents The psychological profile that structurally documents the case of a depression patient is essential for psychotherapy. Large language models can be applied to summarize the profiles from counseling speech, however, it may suffer from long-context forgetting and produce unverifiable hallucinations, due to overlong length of speech, multi-party interactions and unstructured chatting. Hereby, we propose a StreamProfile, a streaming framework that processes counseling speech incrementally, extracts evidences grounded from ASR transcriptions by storing it in a Hierarchical Evidence Memory, and then performs a Chain-of-Thought pipeline according to PM+ psychological intervention for clinical reasoning. The final profile is synthesized strictly from those evidences, making every claim traceable. Experiments on real-world teenager counseling speech have shown that the proposed StreamProfile system can accurately generate the profiles and prevent hallucination.
format Preprint
id arxiv_https___arxiv_org_abs_2604_10161
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle From Speech to Profile: A Protocol-Driven LLM Agent for Psychological Profile Generation
Yang, Xingjian
Yang, Yudong
Guo, Zhixing
Zhou, Yongjie
Yan, Nan
Wang, Lan
Sound
The psychological profile that structurally documents the case of a depression patient is essential for psychotherapy. Large language models can be applied to summarize the profiles from counseling speech, however, it may suffer from long-context forgetting and produce unverifiable hallucinations, due to overlong length of speech, multi-party interactions and unstructured chatting. Hereby, we propose a StreamProfile, a streaming framework that processes counseling speech incrementally, extracts evidences grounded from ASR transcriptions by storing it in a Hierarchical Evidence Memory, and then performs a Chain-of-Thought pipeline according to PM+ psychological intervention for clinical reasoning. The final profile is synthesized strictly from those evidences, making every claim traceable. Experiments on real-world teenager counseling speech have shown that the proposed StreamProfile system can accurately generate the profiles and prevent hallucination.
title From Speech to Profile: A Protocol-Driven LLM Agent for Psychological Profile Generation
topic Sound
url https://arxiv.org/abs/2604.10161