Sharing something I made: Mimi. When watching foreign-language videos, livestreams, or online courses without Chinese subtitles, it can recognize the audio currently playing on your computer and translate it into floating subtitles, without requiring you to download the video first.
The subtitles can be dragged and resized, and you can also switch to immersive mode, which leaves only the subtitles visible. The software is free and open source under the MIT license, supporting Apple silicon Macs running macOS 13+ and Windows x64.
You’ll need to provide your own cloud service API credentials, and providers may charge for their services. It currently mainly uses Alibaba Cloud for testing; other services are also integrated, but haven’t been tested thoroughly and may not work properly.
This is actually a desktop app. An iOS version is being explored, and a similar effect could be achieved with picture-in-picture. The downside is that if the video also enters picture-in-picture mode, this little translation window disappears. Hahaha, I don’t know how to solve that. I’ve also considered some rather unusual approaches, such as using the Dynamic Island to display translated subtitles, but that didn’t work. It would look roughly like this:
I’d like to fulfill a little dream: when I used to watch adult videos, I could only understand the performers’ expressions, so I’d also like to understand a bit of what they were saying ≽(•⩊ •マ≼
