Scholarlink

Modern technology is helping improve education by making classes more interactive and engaging. This research introduces a live speech-to-visual learning system that listens to and transcribes the instructor’s speech using the Whisper neural network, and highlights textual keywords on-screen using spaCy-based keyword extraction. The system employs Streamlit to provide a user interface that updates in real time as the user interacts with it. The goal is to help students better understand and stay engaged by pairing spoken lessons with relevant supporting images. Multiple experiments demonstrate that the system is scalable, operates efficiently, and reliably maps keywords to images. This work offers a promising approach to enhancing classroom learning through AI and multimedia integration.

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

Real-Time Speech-to-Visual Learning System Using Whisper and spaCy for Educational Enhancement

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Deepak Mane,
  • Nimish Somani,
  • Tejas Runwal,
  • Soham Pawar,
  • Aditya Solunke,
  • Tanishka Singh,
  • Purvi Solanki

摘要

Modern technology is helping improve education by making classes more interactive and engaging. This research introduces a live speech-to-visual learning system that listens to and transcribes the instructor’s speech using the Whisper neural network, and highlights textual keywords on-screen using spaCy-based keyword extraction. The system employs Streamlit to provide a user interface that updates in real time as the user interacts with it. The goal is to help students better understand and stay engaged by pairing spoken lessons with relevant supporting images. Multiple experiments demonstrate that the system is scalable, operates efficiently, and reliably maps keywords to images. This work offers a promising approach to enhancing classroom learning through AI and multimedia integration.