Comment by pavon
2 days ago
According to this comment[1] by an OSS developer working on reverse engineering the device, the documentation released doesn't allow them to implement an alternative backend. If I understand the purpose of the interfaces correctly from skimming the reverse engineering effort github[2], the API released documents the HTTP interface between the phone app and the speakers, which has been available for years, and covers functionality that isn't going away. The interface between the speaker and the cloud services that are shutting down is still undocumented.
[1]https://arstechnica.com/gadgets/2026/01/bose-open-sources-it...
A comment here also says this is the case: > Description: Presets are a core part of the SoundTouch ecosystem. A preset is used to set and recall a specific music stream supported by the SoundTouch speaker
There's a GET method that returns information about presets. Presumably you'd use a POST or PUT method to manage the presents. To that end, under POST, it says:
> POST: N/A
It looks like this API basically allows you to control it like an dumb speaker. That's not nothing, but it's not much either.
There doesn't seem to be anything in the API about controlling how the speak communicates with a back end service.
Edit:
Having some time to read over your [2] and the link to [0] it looks like getting root on the speaker w/ physical access is ridiculously easy. Booting the unit w/ a FAT32 USB drive attached with a file named "remote_services" in an otherwise empty root directory opens up an ssh server and the root user has no password.
The comments on [0] have some interesting tidbits in them, too.
These speakers look like they might be fun to play with and once Bose kills the back end people may unload them cheap.
[0]dspillett
1 day ago
Ah, that makes sense as well. Not sure why I fixated on what the speaker might call out to, and didn't think of what might want/need to control the speaker.