Doeviz vs 灵光

Doeviz is an AI assistant designed specifically for people who are blind or have low vision, helping them overcome everyday challenges in life, work, and beyond. By combining AI-powered visual recognition, real-time video interaction, and accessible navigation, Doeviz helps users understand their surroundings, find objects, read text, and navigate more independently, making everyday tasks easier to accomplish without relying on others. Doeviz is designed and developed with active participation from people with visual impairments. The entire app is fully optimized for accessibility and screen readers, ensuring that every feature is intuitive and easy to use. Key Features: 【Video call】 Start a real-time voice conversation with your AI assistant using your camera. The AI continuously observes what the camera sees, answers your questions, helps you find objects, and describes your surroundings as they change. You can interrupt the AI at any time, making conversations feel more natural. 【Accessible Navigation】 Designed around the real-world travel needs of people with visual impairments, with optimized navigation instructions that are more accurate and easier to understand, along with a range of specialized accessibility features. 【Walking Assist】 Uses the phone’s camera and sensors to detect obstacles, steps, and other objects ahead, providing voice guidance about their locations and distances to help users better understand their walking environment. 【Explore】 Ask a question and let the app continuously capture images automatically or manually, providing updated answers based on your original question. 【Pedestrian signals】 Point your phone’s camera at a pedestrian traffic light to automatically recognize its current signal and receive audio and vibration alerts. 【Elevator floors】 Automatically detects floor changes while riding an elevator and notifies you when you reach your destination floor. 【Document reader】 Intelligently guides you to position your phone and scan entire pages of books, documents, and packaging. Extracts and reads text aloud, and lets you ask follow-up questions such as “What page is this?” 【Card scanning】 Captures various types of cards and IDs, briefly describes their contents, and automatically crops the images. 【Instant text】 Quickly finds and reads text on signs, station names, labels, and more, without requiring an internet connection or taking a photo. 【AI Chat】 Talk to your AI assistant, ask questions, and search for information to get help with everyday life, learning, travel, and more. More features designed specifically for people who are blind or have low vision are on the way.

灵光App是蚂蚁集团推出的全模态通用AI助手，包括「灵光对话」、「灵光闪应用」、「灵光开眼」三大功能，是业内首个全代码生成多模态内容的通用AI助手。 作为首个消费级Coding Agent，实现了一句话快速生成可交互可编辑的闪应用，用户也可把自建的闪应用发布到灵光圈，让分享应用像发朋友圈一样简单。 ——核心功能—— ● 灵光闪应用——30秒生成闪应用 【灵光闪应用】一句话生成可交互的小应用。量身定制的答案，随心调节的参数，让每一次好奇发问都能收获个性化解答。无论是运动健身、旅行规划、健康食谱，或是理财规划与趣味互动，生成你的专属应用，在对话中即刻运行，满足美好生活的多元需求。 【灵光圈】业内首个 0 代码手搓应用分享社区。汇聚全网脑洞，海量实用又有趣的闪应用随逛随用。支持点赞、收藏、二次编辑，在互动中解锁更多可能。这里更是用户的灵感舞台，凭创意想法与审美即可生成闪应用并一键发布，让每一个闪应用都有机会成为下一个社区爆款。 ● 灵光对话—— 让知识“活”起来 只需要输入问题或者关键词，灵光即可通过以下形式为你提供可视化的答案； 【语音朗读】无论是英文单词、生僻字发音，还是睡前故事，灵光都能即时转化为自然流畅的语音。 【3D数字模型】运用前沿三维数字化技术，打破平面界限，实现交互式探索，从金字塔到霸王龙，带来穿越时空的沉浸式科普体验。 【生成式插图】让信息拥有专属视觉语言，将“量子纠缠”或“经济学原理”等抽象复杂的知识，化繁为简，直观呈现为图像或动画。 【图表与数据】秒懂复杂数据，财报、研究论文等一目了然。支持追问、深挖从而扩展视角。 【可交互地图】从美食探店到旅行路线，灵光为你直接生成地图，帮你清晰规划路径与动线，查询位置信息。 【解读和溯源】支持在一个对话中针对答案进一步深度解读和查看来源，让理解建立在可靠、透明的知识之上。 ● 灵光开眼—— 你的智能“探索之眼” 灵光的“开眼”功能，赋予你的手机一双能理解、会思考、懂创作的智能探索之眼，支持文生图、文生视频、图生图、图生视频等多种创作玩法。 【实时理解】实时理解复杂场景动态画面，支持语音双向交互问答，无论是路边的植物、博物馆的画作，还是潮流单品，都能即时获得精准解读。 【看图提问】上传图片自由提问，不仅能识别“这是什么”，更能获取“为什么”、“怎么用”等延展知识。 【自由创作】指令修图改图，一句话即可生成视频和图片，为静态照片注入鲜活生命力，解锁自由创作的无限可能。