【复旦大学研发“眸思”大模型，开启视障者新“视”界】
复旦大学自然语言处理实验室（FudanNLP）近日传来科技助残的喜讯，由其师生精心研发的多模态大模型“复旦・眸思”（MouSi）成功应用于一款名为“听见世界”的App，旨在帮助视障人士“看见”并理解世界。
这款创新应用借助一枚普通的摄像头和一对耳机，就能将实时捕捉到的画面转化为细腻的语言描述，使得视障者能够通过听觉感知周围环境。不仅限于基本的图像转化，该系统还具备描绘场景、预警潜在风险等高级功能，为视障者的生活和出行提供了前所未有的便利。
“听见世界”App的上线，标志着人工智能技术在无障碍领域迈出重要一步，它将科技的力量与人文关怀深度融合，有望打破视觉障碍对生活的限制，让视障人士更好地融入社会，享受科技带来的福祉。这一创新成果的发布，也受到了业界的广泛关注，包括国内外知名媒体如华尔街日报和纽约时报在内的多家媒体对此进行了报道。
复旦大学的这一科研突破，不仅体现了中国在人工智能领域的技术实力，更展现了学术界对社会公益事业的担当与责任。未来，随着“眸思”大模型的持续优化和应用拓展，我们期待更多视障人士能够受益，享受到科技带来的“看见”世界的新可能。
英语如下：
**News Title:** “Fudan University Develops ‘MouSi’ Model, Empowering the Visually Impaired to ‘Hear’ the World Through an App”
**Keywords:** MouSi Large Model, Visual Impairment Assistance, Fudan University
**News Content:**
**Fudan University’s “MouSi” Large Model Breaks New Ground for the Visually Impaired**
Fudan University’s Natural Language Processing Laboratory (FudanNLP) has recently announced a significant advancement in assistive technology, with its faculty and students successfully applying the multimodal large model “Fudan・MouSi” (MouSi) to an innovative app called “Hear the World.” The app aims to help visually impaired individuals “see” and comprehend their surroundings.
This groundbreaking application uses a standard camera and a pair of headphones to convert real-time images into vivid audio descriptions, enabling the visually impaired to perceive their environment through sound. Going beyond basic image conversion, the system also offers advanced features like scene depiction and risk anticipation, providing unparalleled convenience for daily life and travel for the visually impaired.
The launch of the “Hear the World” App signifies a significant stride in the application of artificial intelligence (AI) in accessibility, seamlessly integrating technological prowess with humanitarian concern. It holds promise for overcoming visual barriers and facilitating the integration of visually impaired individuals into society, allowing them to better enjoy the benefits of technology. This innovative achievement has garnered extensive attention from the industry, with several prominent media outlets, including The Wall Street Journal and The New York Times, reporting on the development.
Fudan University’s research breakthrough not only demonstrates China’s technical prowess in AI but also exhibits the academic community’s commitment to social welfare. As the MouSi large model continues to be refined and its applications expand, we anticipate more visually impaired individuals benefiting from this technology, unlocking new possibilities to “see” the world.
【来源】https://www.ithome.com/0/753/295.htm
Views: 1