|
Hello everyone!
Since I discovered the CVE-2020-10263 report regarding XiaoAi Pro LX06, I thought there would be lots of possibilities with the speaker, and so I ordered some speakers and begin testing them.
I'm based at Spain, and I can't speak Chinese, so the speaker as-is is useless for me, since I cannot interact with it.
So I'm currently working in a public repository to add new software that allows me to have my own smart speaker, without having to connect to the Cloud Internet Services, and extending functionality, by adding MPD music player, multi-room audio with Snapcast, and more!
As I see this forum has lots of posts regarding Mi Speakers, I thought it would be good to share the repository with you, so you can test it if you want and provide some feedback and improvements.
[/!\] Please be aware that this will patching will remove all MiBrain and related Mi Services, so you will not be able to use the default voice assistant.
Link: https://github.com/duhow/xiaoai-patch
Default patching includes following changes:
- Enable SSH and ssh-keys in /data/ssh, so you can login without password
- Remove mico, MiBrain services
- Allow change Bluetooth name
- Remove Dolby DTS library reference
- Update time NTP after boot finishes
- Update hostname to match the speaker SN, for example LX06-1420 .
- Update SSL CA certificates
- Add custom sounds for different notification sounds
- Custom binaries, volume control, Google TTS
In order to install this patch, you need to extract the rootfs squashfs file - for example on LX06 this is done with dd if=/dev/mtd4 of=/tmp/root.img , then use this as a original file to perform patching.
Once you have build a patched image, apply it on the unused root partition (if you're running on boot0/rootfs0, flash rootfs1), and boot it.
For LX06 I'm currently testing with base image 1.58.15 , newer versions may have different changes and patching may require some additional changes.
Hope you like it, and let me know your opinion. Thanks!
|
评分
-
查看全部评分
|