Scholarlink

ZenMate is a culturally adaptive multilingual mental health chatbot that provides empathetic support by dynamically tailoring its tone and language according to the user's linguistic and cultural context. Built using OpenAI's GPT-3.5 and enhanced with spaCy's language detection capabilities, ZenMate currently supports English, Hindi, and Tamil. Unlike generic chatbots, ZenMate integrates culturally contextualized system prompts that reflect specific social, emotional, and family dynamics relevant to each supported language. This system is implemented using a Flask backend, a user-friendly chat frontend, and features an admin panel for monitoring and logging interactions. The proposed solution bridges the cultural gap in AI therapy by emphasizing language-aware responses and emotional relatability. Evaluation shows increased user satisfaction, improved perceived empathy, and scalable architecture suited for deployment in diverse communities.

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

ZenMate: A Culturally Adaptive Multilingual Mental Health Chatbot Using GPT-3.5 and SpaCy

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Allen Romi,
  • Abhilash P. Rajesh,
  • Safa Mohamed Iqubal,
  • Tekle Negash Hagos

摘要

ZenMate is a culturally adaptive multilingual mental health chatbot that provides empathetic support by dynamically tailoring its tone and language according to the user's linguistic and cultural context. Built using OpenAI's GPT-3.5 and enhanced with spaCy's language detection capabilities, ZenMate currently supports English, Hindi, and Tamil. Unlike generic chatbots, ZenMate integrates culturally contextualized system prompts that reflect specific social, emotional, and family dynamics relevant to each supported language. This system is implemented using a Flask backend, a user-friendly chat frontend, and features an admin panel for monitoring and logging interactions. The proposed solution bridges the cultural gap in AI therapy by emphasizing language-aware responses and emotional relatability. Evaluation shows increased user satisfaction, improved perceived empathy, and scalable architecture suited for deployment in diverse communities.