复旦大学团队成功研发了一款名为“眸思”的大模型，为视障者量身打造了“听见世界”App，帮助他们通过声音感知世界。这款App的上线标志着复旦大学自然语言处理实验室 (FudanNLP) 师生在人工智能领域的又一重要突破。
据复旦大学官方公众号报道，复旦・眸思(MouSi) 是一款基于多模态大模型的应用程序，能够将画面转化为语言，为视障者提供全新的感知体验。这套系统的设计理念源于对视障者需求的深入理解和关注，通过将先进的人工智能技术应用于实际场景，让他们能够更好地融入社会，提高生活质量。
“听见世界”App 的使用非常简便，只需一枚摄像头和一对耳机。用户在佩戴耳机后，App 能够实时将周围的画面转化为语言描述，为视障者提供详细的信息。此外，该系统还具备描绘场景、提示风险等功能，能够在一定程度上模拟正常人的视觉感知，帮助视障者更加安全、自信地行走于世间。
复旦大学团队在研发过程中，始终坚持以人为本的价值观，注重技术的实用性和人性化。他们与视障者进行了深入的交流与合作，根据他们的实际需求进行功能设计和优化。这种以人为本的设计理念，使得“听见世界”App 具有更高的实用价值和用户体验。
此次“复旦・眸思”大模型的成功研发和应用，不仅展示了复旦大学在自然语言处理领域的技术实力，也为我国人工智能事业的发展做出了重要贡献。未来，复旦大学团队将继续关注视障者等特殊群体的需求，用科技的力量为他们创造更多的可能。
这款App的上线，对于视障者来说，无疑是一种全新的感知世界的途径。它让他们能够通过声音感受到周围的环境，听见世界的美好。相信在不久的将来，随着技术的不断进步和创新，人工智能将为他们带来更加丰富和精彩的生活。
英语如下：
# Title: Fudan Team Develops “MouSi” Large Model: Helping the Visually Impaired “See” the World
Keywords: Visual assistance, Fudan University, MouSi Model.
## News Content
A team from Fudan University has successfully developed a large model named “MouSi,” tailor-made for the visually impaired, which powers the “Hear the World” app, helping them perceive the world through sound. The launch of this app marks another significant breakthrough for the Fudan Natural Language Processing Laboratory (FudanNLP) faculty and students in the field of artificial intelligence.
According to an official post from Fudan University’s public account, Fudan・MouSi is an application based on a multimodal large model that can translate images into language, providing the visually impaired with a brand new perception experience. The design concept of this system originates from a deep understanding and attention to the needs of the visually impaired, using advanced artificial intelligence technology to apply practical scenarios, enabling them to better integrate into society and improve their quality of life.
The “Hear the World” app is very easy to use; it only requires a camera and a pair of headphones. After wearing the headphones, the app can translate the surrounding images into language descriptions in real-time, providing the visually impaired with detailed information. In addition, the system also has functions such as sketching scenes and warning of risks, which can simulate a normal person’s visual perception to some extent, helping the visually impaired walk in the world more safely and confidently.
Throughout the research and development process, the Fudan team has always adhered to a people-oriented value proposition, focusing on the practicality and humanization of the technology. They have conducted in-depth exchanges and cooperation with the visually impaired, designing and optimizing functions based on their actual needs. This people-oriented design concept makes the “Hear the World” app have higher practical value and user experience.
The successful development and application of the “Fudan・MouSi” large model not only showcases Fudan’s technical strength in the field of natural language processing but also makes a significant contribution to the development of China’s artificial intelligence industry. In the future, the Fudan team will continue to pay attention to the needs of special groups such as the visually impaired, using the power of technology to create more possibilities for them.
The launch of this app undoubtedly provides the visually impaired with a completely new way to perceive the world. It allows them to feel their surroundings and the beauty of the world through sound. It is believed that with the continuous progress and innovation of technology, artificial intelligence will bring them a more abundant and exciting life in the near future.
【来源】https://www.ithome.com/0/753/295.htm
Views: 1