OpenVoiceOS support for FutureProofHomes Satellite1 HAT Board hardware with Raspberry Pi Zero?

[Hedda]

Any plans to make an image supporting OpenVoiceOS (on Raspberry Pi Zero) with the FutureProofHomes Satellite1 HAT Board hardware?

• https://github.com/orgs/FutureProofHomes/discussions/24

• https://futureproofhomes.net/products/satellite1-pcb-dev-kit

• https://github.com/FutureProofHomes/Satellite1-Hardware

mp4_hero_480_frames.mp4

FutureProofHomes are also working on enclosures and 3D-print enclodure designs;:

• https://futureproofhomes.net/pages/satellite1-enclosure

  • https://docs.futureproofhomes.net/satellite1-dev-kit-enclosure/

    • https://github.com/FutureProofHomes/Satellite1-Enclosures

I understand the FutureProofHomes Satellite1 HAT Board hardware will technically be able use a Raspberry Pi Zero as a compute board but not heard of any plans to develop their own operating-system image themself as they instead pointed to rhasspy's wyoming-satellite repo for those wanting to use Raspberry Pi:

• https://github.com/rhasspy/wyoming-satellite

...as well as working on their own ESPHome firmware image for their ESP32 board based on the Home Assistant Voice PE reference hardware:

• https://github.com/FutureProofHomes/Satellite1-ESPHome
  • https://github.com/esphome/home-assistant-voice-pe

PS: Noticed that FutureProofHomes had a preview video on YouTube mentioning an early prototype of their project as "HomeX" when they at that time has based their prototype on the wyoming-satellite platform running on a Raspberry Pi instead of the released dev kit which instead by default comes with an ESP32 compute board flashed with Nabu Casa's upcoming ESPHome-based voice-kit hardware platform but still using an XMOS xCORE chip for audio processing:

https://www.youtube.com/watch?v=dRTLjQHfjSM

[JarbasAl]

not a huge fan of wyoming protocol, in OVOS we support satellites via the companion project hivemind

the idea has floated around for making a wyoming bridge HiveMindInsiders/feature-tracker#2 and also to look into updating the firmware for speaking native hivemind HiveMindInsiders/feature-tracker#3

nothing concrete is planned, but i would love to see all that hardware working with OVOS!

[Hedda]

I am only asking about supporting the hardware (i.e. the "FutureProofHomes Satellite1 HAT Board"), so not asking about Wyoming support. So just to clarify Wyoming protocol support is totally off-topic and has absolutely nothing directly to do with with this suggestion/question.

[JarbasAl]

wyoming is relevant as mentioned in HiveMindInsiders/feature-tracker#2

the better but more time consuming approach will require updating the firmware to remove wyoming,

a adapter can be made to translate wyoming to Hivemind/OVOS which should in theory allow all the wyoming satellites out there to just work without changing firmware, not as secure and introduces latency etc, but accounts for the users that want to use the stock firmware

[j1nx]

Very interesting device! Will for sure check it out. If it basically rpi zero, I do hae good hope it can be setup as OVOS / Hivemind satelite which opens up another market for the device.

[Hedda]

Note! I am only suggesting to add support for FutureProofHomes Satellite1 HAT Board hardware to the OpenVoiceOS operatingsystem.

As in using the FutureProofHomes Satellite1 HAT Board as a microphone/speaker hardware with a Raspberry Pi Zero running OpenVoiceOS.

Today they only offer it with their own ESP32-based "compute board" but you can replace that with any Raspberry Pi Zero compatible board:

https://futureproofhomes.net/products/satellite1-pcb-dev-kit

Satellite1 HAT Board:

Satellite1 HAT Board features 4 high-performance microphones powered by an XMOS audio processing chip, enabling clear audio capture even in noisy environments. Additionally, the HAT boasts 4 programmable buttons, a headphone jack, and a robust 25W speaker amplifier. It even acts as a multi-sensor by measuring room temperature, humidity, luminosity, and presence when paired with a mmWave radar sensor.

• Powerful Audio Processor: XMOS XU316 with 16-Core 32-Bit RISC running at 800 MHz. 3200MIPs
• Memory: 1024 KB SRAM, 8 MB Flash
• LEDs: 24x WS2812 programmable RGB LEDs
• Microphones: 4x PDM MEMS microphones
• Buttons: 4x Tactile buttons with one hardware controlled mute, for security
• Line level audio output: PCM5122
• 25W Mono speaker output: TAS2780
• USB Power delivery, up to 100W: FUSB302
• Ambient Light sensor: LTR-303ALS
• Temperature and Humidity sensor: AHT20
• Millimeter Wave Radar Presence Detection: Attach your own LD2410 or LD2450
• Expandability: 40-pin GPIO HAT connector, 2 USB Type-C ports, 80-pin accessory connector for upcoming FutureProofHomes mini-speaker, amplifier and in-ceiling mount (more info coming soon).
• Hardware owned by you. No vendor lock-in, or planned obsolescence.
• Power Requirements: The complete assembly requires up to 30W of power at maximum consumption, when using a 20-25W speaker. We recommend at least a 30W power supply, that can deliver 20V @ 1.5A or greater.

[Hedda]

FYI; since it is based on XMOS XU316 advanced DSP microcontroller for audio input and output you can for a DIY setup alternativly also test the similar stand-alone USB-microphone solution with the “ReSpeaker Lite 2-Mic Array” (from Seeed Studio) that can connect direct to a computer:

• https://www.seeedstudio.com/ReSpeaker-Lite-p-5928.html

More in their wiki :

• https://wiki.seeedstudio.com/xiao_respeaker/

and GitHub repo

• https://github.com/respeaker/ReSpeaker_Lite/

Seeed Studio currently sells three different "ReSpeaker Lite" variants and kits containing different components in combination with a board, and note that you can also add speaker and a basic enclosure to each kit to get a stand-alone setup, (as well as a computer like a Raspberry Pi Zero):

All the "ReSpeaker Lite" kits use a solution that is designed to combine an XMOS XU-316 AI (xCORE XU316) which is a dedicated DSP (Digital Signal Processor) Audio Processor Microcontroller IC chipset which can act as a audio co-processor (i.e. onboard sound-card) for all type of advanced on-device sound and voice processing with an ESP32-S3 that can be used to run ESPHome firmware to allow for Voice and Assist integration with Home Assistant integration, and they have a wiki with instructions on how to set it up (though unclear if and how that setup is utilized the XMOS chip for advanced audio processing):

The XMOS xCORE DSP chip acts like sound-card co-processor adding in-line off-loading of audio noise removal (voice clean-up) from the microphone(s), like Interference Cancellation (IC)​, Acoustic Echo Cancellation (AEC), Noise Suppression (NS), and Automatic Gain Control, etc. and/and other audio post-processing algorithms to improve the solution’s voice recognition capabilities). (Depending on which XMOS chip they use their XCORE-VOICE framework could technically also allow also for up to 16 PDM microphones to be connected to a single xCORE device with a different PCB design).

• https://www.xmos.com/develop/xcore-voice

Note that these “ReSpeaker Lite” kits (all three available via the same webpage) seem to specifically be targeting ESPHome voice-assistant developers and early-adopter tinkerers, and currently looks to be available as a full development-kit that including an ESP32-S3 based board (Seeed Studio’s XIAO ESP32S3) with simple acrylic enclosure and a speaker or a smaller DIY-kit that only includes the ReSpeaker Lite board where to user will needs to add their own MCU board, SBC or other computer using I2S and/or USB connections, as well as a larger third-kit that include both those two kits. Both the boards have on-board dual-microphone array (2-Mic Array) for far-field voice control interaction, and it says that these kits support custom firmware updates via DFU-Util (though unclear if that includes firmware and/or configuration for the XMOS as well), and it also features an stereo audio output jack for connecting external (Hi-Fi) speakers.

Seeed Studio published these videos showing the stand-alone ESP32 model and the USB model:

https://www.youtube.com/watch?v=C2pXledr8kM

https://www.youtube.com/watch?v=I9KOY2ik5nw

PS: CNX Software has a nice summary blog article with details on the technical hardware specification:

https://www.cnx-software.com/2024/08/05/respeaker-lite-voice-assistant-kit-combines-xmos-xu-316-and-esp32-s3-for-advanced-voice-processing-home-assistant-integration/

[Hedda]

Tip regarding "ReSpeaker Lite" is to check out the 3D-print enclodure from the formatBCE's Koala Satellite project based on Respeaker Lite:

• https://github.com/formatBCE/Koala-Satellite

  • https://github.com/formatBCE/Koala-Satellite/blob/main/instructions/assembly.md

    • https://github.com/formatBCE/Koala-Satellite/blob/main/instructions/parts_and_tools.md

Not sure if it would fit a Raspberry Pi Zero out-of-the-box but might be able to modify the design to fit a Raspberry Pi Zero instead of ESP32?

https://github.com/formatBCE/Koala-Satellite/tree/main/casing

Parts and tools needed are:

Parts used

• microphones and DAC/AMP: Respeaker Lite board (bare board, without XIAO-S3 soldered on, like this);
• main board: ESP32-S3 DevkitC N16R8 like this;
• rotary encoder: KY-040 (i used this);
• LED: 144 diodes/meter addressable WS2812B, 43 diodes long (i used this);
• speaker: 2" 4Ohm 5W (example);
• 10x M3 screws (i used this set);
• 10x threaded inserts (i used this);
• Jst 2.0 Ph 2-Pin Connector (male and female) for speaker like this;
• some 3-wired male-female connector for easy LED strip connection (i used one from this kit);
• cords of various length;
• 4x sticky rubber pads (to isolate enclosure vibrations from surface it's standing on).

Tools used

• 3D printer for enclosure printing (better is better, i use Bambulab A1);
• prints cleanup tool;
• soldering iron and accessories;
• threaded inserts installation tool;
• screwdriver or allen wrench for screws.

[j1nx]

The specifications look great and we do know the XMOS chips. The SJ-201 HAT used within the Mycroft Mark2 device is based on their (now rather old) VocalFusion 3510 INT chip for which I wrote the linux driver for the newer linux kernel a while back.
https://github.com/OpenVoiceOS/VocalFusionDriver

Looks like I have some research, reading and digging to do the upcoming days.

[Hedda]

From my own research I believe that they are all using the XU316-1024-QF60B family of chips, as that chip is native 3.3v IO (which is same as an ESP32), specifically the ”XU316-1024-QF60B-C32" SKU of the XMOS xCORE AI chips because the "C32" models offer 3200 MIPS in performance compared to 2400 MIPS of "C24" models.

https://www.xmos.com/file/xu316-1024-qf60a-xcore_ai-datasheet?version=latest

https://www.xmos.com/download/XU316-1024-QF60A-xcore_ai-Datasheet(26).pdf

A good start could also be to take a look at the XMOS's "XK-VOICE-L71 Voice Reference Design Evaluation Kit" as almost all of these projects are using as a reference design because it is very well documented and tested:

https://www.xmos.com/xk-voice-l71

https://www.xmos.com/file/xu316-1024-qf60a-datasheet/?version=latest

https://www.xmos.com/file/xk-voice-l71-hardware-manual/?version=latest

https://www.xmos.com/file/xk-voice-l71-pcb-design-files/?version=latest

https://www.xmos.com/download/XVF3610-User-Guide(v5_7_3).pdf

https://www.xmos.com/develop/xcore-voice