Setup Virtual Environment

If you are installing on the Bookworm version of Raspberry Pi OS or later, you will need to install your python modules in a virtual environment. You can find more information in the Python Virtual Environment Usage on Raspberry Pi guide. To Install and activate the virtual environment, use the following commands:

sudo apt install python3-venv
python -m venv env --system-site-packages

sudo apt install python3-venv
python -m venv env --system-site-packages

To activate the virtual environment:

source env/bin/activate

source env/bin/activate

Installer script

Luckily its quite easy to install support for I2S DACs on Raspbian.

These instructions are totally cribbed from the PhatDAC instructions at the lovely folks at Pimoroni!

Run the following from your Raspberry Pi with Internet connectivity:

sudo apt install -y wget
pip3 install adafruit-python-shell
wget https://github.com/adafruit/Raspberry-Pi-Installer-Scripts/raw/main/i2samp.py
sudo -E env PATH=$PATH python3 i2samp.py

sudo apt install -y wget
pip3 install adafruit-python-shell
wget https://github.com/adafruit/Raspberry-Pi-Installer-Scripts/raw/main/i2samp.py
sudo -E env PATH=$PATH python3 i2samp.py

We've added an extra helper systemd script that will play quiet audio when the I2S peripheral isn't in use. This removes popping when playback starts or stops. It uses a tiny amount of CPU time (on a Pi Zero, 5%, on a Pi 2 or 3 its negligible). You don't need this on RetroPie because it never releases the I2S device, but it's great for Raspbian.

You will need to reboot once installed.

After rebooting, log back in and re-run the script again...It will ask you if you want to test the speaker. Say yes and listen for audio to come out of your speakers...

If it sounds really distorted, it could be the volume is too high. However, in order to have volume control appear in Raspbian desktop or Retropie you must reboot a second time after doing the speaker test, with sudo reboot

Once rebooted, try running alsamixer and use arrow keys to lower the volume, 50% is a good place to start.

If you're still having audio problems, try re-running the script and saying N (disable) the /dev/zero playback service.

You can then go to the next page on testing and optimizing your setup. Skip the rest of this page on Detailed Installation if the script worked for you!

Detailed Install

If, for some reason, you can't just run the script and you want to go through the install by hand - here's all the steps!

Update /etc/modprobe.d (if it exists)

Log into your Pi and get into a serial console (either via a console cable, the TV console, RXVT, or what have you)

Edit the raspi blacklist with

sudo nano /etc/modprobe.d/raspi-blacklist.conf

sudo nano /etc/modprobe.d/raspi-blacklist.conf

If the file is empty, just skip this step

However, if you see the following lines:

blacklist i2c-bcm2708
blacklist snd-soc-pcm512x
blacklist snd-soc-wm8804

Update the lines by putting a # before each line

Save by typing Control-X Y <return>

Disable headphone audio (if it's set)

Edit the raspi modules list with

sudo nano /etc/modules

sudo nano /etc/modules

If the file is empty, just skip this step

However, if you see the following line:

snd_bcm2835

Put a # in front of it

and save with Control-X Y <return>

Create asound.conf file

Edit the raspi modules list with

sudo nano /etc/asound.conf

sudo nano /etc/asound.conf

This file ought to be blank!

Copy and paste the following text into the file

pcm.speakerbonnet {
   type hw card 0
}

pcm.dmixer {
   type dmix
   ipc_key 1024
   ipc_perm 0666
   slave {
     pcm "speakerbonnet"
     period_time 0
     period_size 1024
     buffer_size 8192
     rate 44100
     channels 2
   }
}

ctl.dmixer {
    type hw card 0
}

pcm.softvol {
    type softvol
    slave.pcm "dmixer"
    control.name "PCM"
    control.card 0
}

ctl.softvol {
    type hw card 0
}

pcm.!default {
    type             plug
    slave.pcm       "softvol"
}

pcm.speakerbonnet {
   type hw card 0
}

pcm.dmixer {
   type dmix
   ipc_key 1024
   ipc_perm 0666
   slave {
     pcm "speakerbonnet"
     period_time 0
     period_size 1024
     buffer_size 8192
     rate 44100
     channels 2
   }
}

ctl.dmixer {
    type hw card 0
}

pcm.softvol {
    type softvol
    slave.pcm "dmixer"
    control.name "PCM"
    control.card 0
}

ctl.softvol {
    type hw card 0
}

pcm.!default {
    type             plug
    slave.pcm       "softvol"
}

Save the file as usual

Add Device Tree Overlay

 Edit your Pi configuration file with

sudo nano /boot/firmware/config.txt

sudo nano /boot/firmware/config.txt

And scroll down to the bottom. If you see a line that says: dtparam=audio=on

Disable it by putting a # in front.

Then add:

dtoverlay=max98357a

dtoverlay=max98357a

on the next line. Save the file and reboot your Pi with

sudo reboot

sudo reboot

Speaker Tests!

OK you can use whatever software you like to play audio but if you'd like to test the speaker output, here's some quick commands that will let you verify your amp and speaker are working as they should!

Simple white noise speaker test

Run

speaker-test -c2

speaker-test -c2

to generate white noise out of the speaker, alternating left and right.

If you have a mono output amplifier, the I2S amp merges left and right channels, so you'll hear continuous white noise

Simple WAV speaker test

Once you've got something coming out, try to play an audio file with speaker-test (for WAV files, not MP3)

speaker-test -c2 --test=wav -w /usr/share/sounds/alsa/Front_Center.wav

speaker-test -c2 --test=wav -w /usr/share/sounds/alsa/Front_Center.wav

You'll hear audio coming from left and right alternating speakers

Simple MP3 speaker test

If you want to play a stream of music, you can try

sudo apt-get install -y mpg123
mpg123 http://ice1.somafm.com/u80s-128-mp3

sudo apt-get install -y mpg123
mpg123 http://ice1.somafm.com/u80s-128-mp3

If you want to play MP3's on command, check out this tutorial which covers how to set that up

At this time, Jessie Raspbery Pi kernel does not support mono audio out of the I2S interface, you can only play stereo, so any mono audio files may need conversion to stereo!

Volume adjustment

Many programs like PyGame and Sonic Pi have volume control within the application. For other programs you can set the volume using the command line tool called alsamixer. Just type alsamixer in and then use the up/down arrows to set the volume. Press Escape once its set

In Raspbian PIXEL you can set the volume using the menu item control. If it has an X through it, try restarting the Pi (you have to restart twice after install to get PIXEL to recognize the volume control

Reducing popping

For people who followed our original installation instructions with the simple alsa config, they may find that the I2S audio pops when playing new audio.

The workaround is to use a software mixer to output a fixed sample rate to the I2S device so the bit clock does not change. I use ALSA so I configured dmixer and I no longer have any pops or clicks. Note that the RaspPi I2S driver does not support dmixer by default and you must follow these instructions provided to add it. Continue on for step-by-step on how to enable it!

Step 1

Start by modify /boot/firmware/config.txt to adddtoverlay=i2s-mmap

Run sudo nano /boot/firmware/config.txt and add the text to the bottom like so:

Save and exit.

Then change /etc/asound.conf to:

pcm.speakerbonnet {
   type hw card 0 
}

pcm.!default { 
   type plug 
   slave.pcm "dmixer" 
}

pcm.dmixer { 
   type dmix 
   ipc_key 1024
   ipc_perm 0666
   slave { 
     pcm "speakerbonnet" 
     period_time 0
     period_size 1024
     buffer_size 8192
     rate 44100
     channels 2 
   } 
}

ctl.dmixer { 
  type hw card 0 
}

pcm.speakerbonnet {
   type hw card 0 
}

pcm.!default { 
   type plug 
   slave.pcm "dmixer" 
}

pcm.dmixer { 
   type dmix 
   ipc_key 1024
   ipc_perm 0666
   slave { 
     pcm "speakerbonnet" 
     period_time 0
     period_size 1024
     buffer_size 8192
     rate 44100
     channels 2 
   } 
}

ctl.dmixer { 
  type hw card 0 
}

By running sudo nano /etc/asound.conf

This creates a PCM device called speakerbonnet which is connected to the hardware I2S device. Then we make a new 'dmix' device (type dmix) called pcm.dmixer. We give it a unique Inter Process Communication key (ipc_key 1024) and permissions that are world-read-writeable (ipc_perm 0666) The mixer will control the hardware pcm device speakerbonnet (pcm "speakerbonnet") and has a buffer set up so its nice and fast. The communication buffer is set up so there's no delays (period_time 0, period_size 1024 and buffer_size 8192 work well). The default mixed rate is 44.1khz stereo (rate 44100 channels 2)

Finally we set up a control interface but it ended up working best to just put in the hardware device here -ctl.dmixer { type hw card 0 }

Save and exit. Then reboot the Pi to enable the mixer. Also, while it will greatly reduce popping, you still may get one once in a while - especially when first playing audio!

Add software volume control

The basic I2S chipset used here does not have software control built in. So we have to 'trick' the Pi into creating a software volume control. Luckily, its not hard once you know how to do it.

Create a new audio config file in ~/.asoundrc with nano ~/.asoundrc and inside put the following text:

pcm.speakerbonnet {
   type hw card 0
}

pcm.dmixer {
   type dmix
   ipc_key 1024
   ipc_perm 0666
   slave {
     pcm "speakerbonnet"
     period_time 0
     period_size 1024
     buffer_size 8192
     rate 44100
     channels 2
   }
}

ctl.dmixer {
    type hw card 0
}

pcm.softvol {
    type softvol
    slave.pcm "dmixer"
    control.name "PCM"
    control.card 0
}

ctl.softvol {
    type hw card 0
}

pcm.!default {
    type             plug
    slave.pcm       "softvol"
}

pcm.speakerbonnet {
   type hw card 0
}

pcm.dmixer {
   type dmix
   ipc_key 1024
   ipc_perm 0666
   slave {
     pcm "speakerbonnet"
     period_time 0
     period_size 1024
     buffer_size 8192
     rate 44100
     channels 2
   }
}

ctl.dmixer {
    type hw card 0
}

pcm.softvol {
    type softvol
    slave.pcm "dmixer"
    control.name "PCM"
    control.card 0
}

ctl.softvol {
    type hw card 0
}

pcm.!default {
    type             plug
    slave.pcm       "softvol"
}

Save and exit

Now, here's the trick, you have to reboot, then play some audio through alsa, then reboot to get the alsamixer to sync up right:

speaker-test -c2 --test=wav -w /usr/share/sounds/alsa/Front_Center.wav

Then you can type alsamixer to control the volume with the 'classic' alsa mixing interface

Just press the up and down arrows to set the volume, and ESC to quit