ooboqoo/shiping
识屏 Shiping:截屏 + AI,看懂屏幕上的任何内容。代码、报错、题目、图表、外文资料,按快捷键截一下就有解答。对话模式实时听懂问题、给出回答要点。可用作面试助手、考试助手、刷题助手。
What it solves
It provides real-time AI assistance for on-screen content and live audio conversations. It helps users quickly understand technical problems, code errors, charts, or foreign language documents without needing to copy-paste or switch windows, and offers real-time prompts during voice-based interactions like interviews.
How it works
It operates in two modes:
- Screenshot Mode: Captures a portion of your screen and sends it to a vision-capable AI model to provide immediate answers or code solutions.
- Dialogue Mode: Uses real-time speech-to-text (via Alibaba Cloud Bailian) to listen to ongoing conversations and generates concise, actionable prompts or talking points for the user.
Both modes are highly customizable, allowing users to select specific AI models, define custom prompts, and even reference local documents (PDFs, Markdown, etc.) to ground the AI's responses.
Who it’s for
It is designed for professionals and students who need quick, unobtrusive AI support during technical interviews, online exams, or complex coding tasks. It includes features like "stealth mode" to keep the assistant window hidden during screen sharing and mobile synchronization for viewing answers on a separate device.
Highlights
- Dual-Mode Assistance: Seamlessly switch between visual screen analysis and audio-based conversation prompts.
- Privacy-Focused: Keeps the assistant window hidden from screen-sharing participants and supports local document referencing.
- Flexible AI Integration: Compatible with any OpenAI-compatible API, allowing users to choose their preferred providers (e.g., DeepSeek, SiliconFlow).
- Optimized for Speed: Supports disabling "thinking" processes in models for faster responses and provides real-time, non-intrusive UI overlays.
- Mobile Sync: View AI responses on a smartphone via local network or public relay to keep the primary computer screen clear.
Related
- Project
- Project
- Project
- Project