Scholarlink

This paper presents the development of an emotionally intelligent robot system that integrates a language model (LLM) with computer vision–based emotion recognition to support basic healthcare interactions. The main objective was to design an empathetic conversational assistant capable of interpreting the user’s emotional state and generating contextually appropriate verbal responses. The proposed system combines three main components: (1) a language-based voice assistant, (2) an emotion recognition module implemented with pretrained Python models, and (3) a multimodal interface running on a Raspberry Pi equipped with camera, microphone, and speakers. The implementation employs containerized environments and lightweight frameworks to ensure modularity, scalability, and local data processing. The integration of both components demonstrates the technical feasibility of creating low-cost, locally operated robots with emotional awareness. This work contributes to the field of affective computing by providing an accessible and adaptable framework for developing emotion-aware healthcare assistants.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Proposal Model to Design an Emotional Assistance Robot to First Assistance at Health Area

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Hector Cruz-Martinez,
  • Leticia Neira-Tovar,
  • Gabriel Monroy

摘要

This paper presents the development of an emotionally intelligent robot system that integrates a language model (LLM) with computer vision–based emotion recognition to support basic healthcare interactions. The main objective was to design an empathetic conversational assistant capable of interpreting the user’s emotional state and generating contextually appropriate verbal responses. The proposed system combines three main components: (1) a language-based voice assistant, (2) an emotion recognition module implemented with pretrained Python models, and (3) a multimodal interface running on a Raspberry Pi equipped with camera, microphone, and speakers. The implementation employs containerized environments and lightweight frameworks to ensure modularity, scalability, and local data processing. The integration of both components demonstrates the technical feasibility of creating low-cost, locally operated robots with emotional awareness. This work contributes to the field of affective computing by providing an accessible and adaptable framework for developing emotion-aware healthcare assistants.