AI Thinker ESP32 Audio Snapcast Multiroom Network Speaker

This is a case for an AI Thinker ESP32 Audio development boards plus a single 53mm speaker. Assembled and flashed with a suitable firmware, it can be used as a node in a multiroom speaker setup based on Snapcast. 

What you need:

• the printed case

• 12 brass 4.5mm M3 inserts

• 12 M3 screws

• a 53mm speaker that has ~35mm depth (I used these)

• an AI Thinker ESP32 Audio board (also on Aliexpress)

• a JST 2.54 XH 2 Pin connector with wire (Amazon).

What you need to do:

• print the case components.

• melt the brass inserts into the case (be careful to keep them straight, small tolerances).

• solder the JST connector wire to the speaker.

• screw the speaker into the top case.

• connect the speaker to one of the channels on the board.

• screw the bottom case to the top case.

• finally, screw the grill on the top case.

• go to the Snapcast Web Installer and select "ESP32-Audio-kit (ES8388)". Install the firmware and configure the Wifi.

Some notes:

• The auto discovery for Snapcast did not work for me, not sure if this was because my Snapcast server is in the wired network and the speakers use the Wifi. If you can't find the speaker on your Snapcast service, you can configure a static server in the web settings (the we installer will give instructions for it).

• If you don't get any audio, play around with the channel settings in your Snapcast server configuration, I had to switch to explicitly mono output on the connected channel to make it work.

• As a Snapcast server, Music Assistant is a great choice. It can play from Spotify and other services and is integrated into your Home Assistant smart home.

• If you run it without the case, the sound quality will probably suck. The case makes a significant difference in the quality. Put it together :-)

• There is an equalizer built into the web settings of the Snapcast firmware.