I accidentally pressed one of the buttons on my respeaker lite while I was moving it and got an factory reset message. The message went on “press button again to confirm”. I didn’t want a factory reset so I didn’t press the button again. However, the respeaker was unresponsive so I powered it off and powered it on again. After restarting, it is now responding normally. However, the blue LED on top is flashing fast, which I think indicates it’s performing some set up or configuration.
I’m running @formatBCE yaml and just wondering what I can do, other than functional testing to ensre it’s working okay?
It has been on for nearly an hour now. The blue LED is flashing fast and the device is responding normally
Probably something went wrong with internal S3 flash memory. In PE (and Respeaker) there’s no confirming by pressing it again - rather it says “keep pressing button to reset”. So i’d say you did factory-reset the device.
Try deleting the device from HA and re-adding it. If it doesn’t help, re-flash.
I’m curious how to raise visibility of Respeaker related information.
Many people struggle to replace their Alexa/Google speakers with Voice PE now. The speaker on PE isn’t good, so there’s only 3.5mm option, or disassembling and using different speaker.
Same stuff happens with FPH SAT1, it’s not ready yet to replace proprietary satellites (although it’s moving there).
On the other hand, Respeaker Lite Satellite and Koala are the closest to PE on the functionality, and have pretty good speaker and look. I try to help people on FB and Discord, but still if I try to search Google for home assistant satellites (or even directly by “Koala satellite home assistant” and "Respeaker Lite home assistant), there’s no good results even on these specific requests.
Do you think it would be possible to use the “ReSpeaker Lite 2-Mic Array” USB board without ESP32 (or alternativly the FutureProofHomes Satellite1 Dev Kit) in combination a ESP32-S3 development board and one of these 7-inch series of capacitive touchscreen displays from Elecrow in order to make your own variant of a Home Assistant Voice Satellite with a screen? …i.e. something similar in design to Google Nest Hub and Amazon Echo Show
At least these “CrowPanel Advance” display series features “replaceable wireless modules” meaning that the ESP32 module can be replaced. That is, this article says that they feature an ESP32-S3-WROOM-1-N16R8 module soldered on the back + having a pin header taking ESP32-H2, ESP32-C6 or…
The specificiation does say that it also features a built-in microphone (however I assume that is rubbish) as well as a speaker connector (but no mention of a DAC).
The also have loads of different models with various sizes of screens. With the “Advanced” series all having a pre-soldered ESP32-S3-WROOM-1-N16R8 module:
CrowPanel Advance ESP32 displays are available on Aliexpress from $25.99 up to $54.19 depending on the selected model and whether any of the wireless modules shown above is added as an option. But you can also purchase the displays and wireless modules on Elecrow’s company’s store. Again see:
The surcharge for the second module is also comparable to the surcharge for a Respeaker Lite WITH its own Xiao, so not sure why you would bother with all the complications…
I have augmented my PR to add a volume control for the respeaker lite with embedded esp s3 that uses the koala case. https://github.com/rananna/Koala-Satellite/blob/fde433be052122879c4d05a31108ccac3a2bae77/README.md
The D1 mini is inside the koala case, and communicates with the koala media controller via home assistant.
This allows the respeaker lite with embedded esp s3 to have a volume control!
An additional benefit is the encoder switch actions have been added to home assistant as events that can be included in automations.
The button appears to be very stable and the D1 mini does not appear to have the debouncing issues that formatbce reported with using the external ESP s3.
Also the encoder is VERY responsive - no lag noticed.
I am experiencing a somewhat random issue. After some time, (could be days or several hours) koala will no longer recognize the wake word. (Ok nabu) .if I reboot koala, it starts working again. Anybody else seeing this?.
No, but i have another thing - if it’s sitting for long, and i try to play music, sometimes it reboots itself. After that, music is playing without troubles. Looks like there’s memory leak somewhere. I’m not that experienced to dig into PE firmware (which we’re using), so hopefully Nabu devs will fix that - and i’ll snatch that.
I meant if they second ESP32 could be connected via SPI kind of like the Espressif’s ESP Thread Border Router SDK hardware (based on ESP32-H2-MINI-1 module):
Personally I would rather use the P4’s dedicated LAN pins and use Ethernet simply due to latency and maybe a bit more bandwidth. The ESP32-S3 has about 256kbps bandwidth. You might get a spike above 500kbps at first but it quickly drops down to 200 to 250kbps. This is what’s documented and matches the speeds I get when sending a file to my P1 via SFTP so no cloud involved. I don’t think this would help with the speech recognition part as if 16000Khz is enough then I doubt any added bandwidth would help but audio playback over JST would more then likely benefit as even Spotify is like 350 to 400kbps.
I ended up getting an Nvidia Jetson Orin NX 16GB and using the whisper/piper/OpenWakeWord (not sure if it’s needed at this point) that Nvidia and Nabu worked on together is night and day compared to my Intel NUC. I’ve been meaning to try Assist Microphone on the Jetson Along with a small LLM models like llama3.2b it works great for general questions, especially with the newish “fallback for local command” option. The problem with an LLM completely controlling Home Assistant is it requires a much larger model, at least until someone creates once specifically for HA which I have no doubt will happen but I also have no idea on the timetable. I’m not sure if the new way cheaper 8GB model, which is just an Orin NX 8GB with a new name would work nearly as good. I’m already over 8GB of RAM after one question so at that point you would be on swap and that will slow things down significantly. If I would of had a motherboard and other parts I would have just bought a GPU but once your at the point were outside maybe a case and power supply is all I have the Jetson was the cheaper way to go.
The one thing that the respeaker lite and HA PE have that I wouldn’t want to not have with an LLM is a stop word. I honestly had good luck with the Wyoming Satellite but it was a stop gap measure until Nabu got it working on ESP32. I highly doubt you will see any future improvements but I could be wrong. Also, even with noise and Echo cancelation they just don’t work well around TV’s, especially something like the news so I created 3 template sensors for the assist_satellite phases. Then created 2 automations. The first takes a scene snapshot of my soundbar and lowers it to a low enough volume where voice commands always work. The second automation restores the snapshot when it goes from listening to processing or processing to replaying as local commands tend to go so fast they don’t catch the processing phase but LLM questions do. Mute works also if you can’t control the actual sound level on the audio device. For anyone wanting to make their own JST speaker connector Minor soldiering skills required.
I made same automation, but put it into one instead of two, and made that generic (i have 5 satellites in different areas, and each of them is lowering the volume on playing players in corresponding area (except of itself, of course, because i want to hear an answer and it ducks active audio anyways).
works pretty good most of the time.
