图书馆杂志

图书馆杂志 ›› 2026, Vol. 45 ›› Issue (5): 72-81.

• 全民阅读推广学坛 • 上一篇    下一篇

从文本到视听:AIGC助力古诗古曲多模态阅读推广的技术路径研究

唐绮蔚,孔馨仪,张嘉慧,冯卓彤,陈涛   


  • 出版日期:2026-05-15 发布日期:2026-05-27
  • 作者简介:唐绮蔚 中山大学信息管理学院,本科生。研究方向:数字人文。作者贡献:AIGC生成、实验方案设计与实现、论文撰写。E-mail:tangqw9@mail2.sysu.edu.cn  广东广州 510006 
    孔馨仪 中国音乐学院音乐学系,硕士研究生。研究方向:民族音乐学、中国传统音乐。作者贡献:文化阐释、论文撰写。北京 100101
    张嘉慧 中山大学信息管理学院,本科生。研究方向:人工智能。作者贡献:AIGC生成,实验方案设计、撰写与实现,论文撰写。广东广州 510006
    冯卓彤 中山大学信息管理学院,硕士研究生。研究方向:数字人文、文化遗产。作者贡献:文献调研、论文撰写。广东广州 510006
    陈涛 中山大学信息管理学院,副教授,硕士生导师。研究方向:数字人文、文化遗产。作者贡献:选题策划、写作指导。广东广州 510006

From Text to Audio Visual: Research on the Technical Pathway of AIGC-Assisted Multimodal Reading Promotion for Ancient Poetry and Music

Tang Qiwei, Kong Xinyi, Zhang Jiahui, Feng Zhuotong, Chen Tao   

  • Online:2026-05-15 Published:2026-05-27
  • About author:Tang Qiwei, Kong Xinyi, Zhang Jiahui, Feng Zhuotong, Chen Tao

摘要: 在数字化与智能化深度融合的背景下,传统阅读推广面临形式单一、传播受限等挑战,而中华优秀传统文化的传承亟须创新路径。本文以《春江花月夜》古诗古曲为研究对象,探索人工智能生成内容(AIGC)技术助力多模态阅读推广的技术路径与实践效果。通过构建《春江遗梦曲承传》视听作品,研究提出“文本—图像—视频”的多模态转化框架,结合AIGC图片生成、视频生成技术场景及提示词工程等技术实践,实现古诗场景的动态可视化与古曲情感的视听联觉表达。研究发现,AIGC技术能够突破传统推广的时空限制,通过多感官体验增强用户沉浸感,促进文化资源的活化与传播;同时,其个性化创作模式降低了专业门槛,激发公众参与文化生产的热情,在提升阅读兴趣与传播效能方面具有显著优势,为数字时代文化传承提供了创新思路。

关键词: 阅读推广, AIGC, 视听联觉, 多模态, 数字人文

Abstract: Against the backdrop of deep integration of digitalization and intelligence, traditional reading promotion faces challenges such as monotonous formats and limited dissemination, while the preservation and revitalization of China's outstanding cultural heritage urgently calls for innovative approaches. This paper takes the ancient poem and musical composition A Moonlit Night on the Spring River as the research subject, and explores the technical pathways and practical impacts of generative artificial intelligence (AIGC) in enabling multimodal reading promotion. By constructing the audio-visual project Spring River Legacy: Inheritance of Ancient Melodies, the study proposes a “text-image-video” multimodal transformation framework. Through techniques such as AIGC image generation, video generation scenarios, and prompt engineering, the research achieves dynamic visualization of ancient poetic scenes and audio-visual synesthetic expression of musical emotions. The research finds that AIGC technology can break through the spatiotemporal constraints of traditional reading promotion, enhance user immersion through multisensory experiences, and facilitate the revitalization and dissemination of cultural resources. Simultaneously, its personalized creation model lowers professional barriers, stimulates public enthusiasm for cultural production, and demonstrates significant advantages in boosting reading interest and communication efficiency, offering innovative insights for cultural transmission and inheritance in the digital age.