Free Your Hands: AI Haoji Automatically Converts Videos into Notes and Mind Maps
Currently, there are many AI programs for text processing on the market, but few for AI organization and summarization of video or audio. Until recently, xiaoz discovered a tool developed by a Tsinghua team called AI Haoji. It not only accurately converts video speech to text but also organizes it into beautiful text-image notes, generates mind maps, and summarizes video content with AI.
Effect as shown below:
What is AI Haoji?
Simply put, it is a multimodal AI knowledge base. Unlike market tools with single functions, it supports both APP and website ends.
The core logic is: put audio and video into it (whether online links or local files), let AI process them, and transform them into structured notes, mind maps, or even podcasts to listen to.
Test Results
1. Direct Link Parsing
Links from Bilibili, Xiaohongshu, Douyin, and Xiaoyuzhou can be parsed directly by copying and pasting. xiaoz tested a 40-minute Bilibili science video, and the parsing speed was quite fast.
2. Local File Upload Parsing
In addition to online links, it supports uploading local audio and video files. Even large files up to 4GB or 4 hours long can be processed stably without pressure.
3. Linkage with Baidu/Aliyun Cloud Drive Upload
Feature Experience: More Than Just Transcription
1. Immersive Notes and PPT Extraction
Many videos have PPT demonstrations, so reading text alone is not enough. AI Haoji automatically captures video frames during transcription and inserts them into the notes.
It also supports timestamps and speaker identification: xiaoz tested a podcast with three people, and it accurately distinguished different speakers. You can double-click to change names. It handles raw videos without subtitles well. However, in some cases, there may be typos, especially in mixed Chinese and English, which needs improvement by the product team.
2. Lazy Person's Gospel, Generate Mind Maps
Video too long to watch? Generate a mind map with one click! The best part is: on the website, clicking a mind map node jumps the video progress bar to the corresponding position. Click where you don't understand, and review efficiency soars.
3. AI Learning Assistant
This is the most black-tech-like feature! It is built with the DeepSeek R1 model. You can chat directly based on audio and video content. For example, upload a video on how to do SEO, then select [Quick Review], and it will generate a complete review document.
4. AI Podcast: Listen to the Video
You can convert boring long videos into a two-person dialogue podcast. There are options for mature female voices and youthful voices. On the way to work or doing housework, turn the long video you used to watch into a podcast to listen to, which is eye-friendly and utilizes fragmented time.
Difference Between APP and Website
- APP: Suitable for recording anytime, parsing links, or listening to AI podcasts during commute
- Website: Suitable for deep organization, batch upload, editing mind maps, and exporting notes
xiaoz suggests using them together.
Benefits Time
Although it is commercial software, the current promotion period benefits are very good:
- New users get 120 points upon registration
- Must fill in the invitation code! Enter the invitation code during registration to get an extra 60 points, totaling 180 points, enough to parse several hours of video. Don't miss out!
How to Get AI Haoji
- APP & Web End Universal
- Exclusive Registration Link: https://aihaoji.com/zh?utm_source=invite&utm_content=Lgz3oa79
- Exclusive Invitation Code:
Lgz3oa79, filling in the invitation code gets an extra 60 points!
Conclusion
In the past, taking notes while watching videos was exhausting. Now AI Haoji handles it directly for me - transcribing text, generating mind maps, listening as a podcast, and even automatically capturing PPTs. Are you sure you don't want to try it?