Have an issue? Look here for support


General help

Booting and configuring

The Android app can’t pair with my Vision Kit or Voice Kit

There are a few reasons why this might happen. First, make sure your Voice Kit is still connected to a power supply and that the green LED on your Bonnet is still flashing. If it’s not flashing, it may have timed out. Press and hold the Bonnet button for 5 seconds, and try again.

If that doesn’t work, first try restarting the app. Then if that doesn’t work, your phone could be experiencing issues with Bluetooth, so try restarting your phone and starting the app pairing flow again.

I want to SSH into my AIY kit

The system images for the Vision Kit, Voice Kit, and Maker Kit have SSH enabled by default. As long as you know the IP address for your Raspberry Pi, you should be able to SSH from a terminal on your personal computer.

For instructions, see the Raspberry Pi docs about remote access.

I’ve tried everything and my kit still won’t boot

If you’ve tried all of the other suggested troubleshooting steps and still can’t get your kit to boot, your SD card might be corrupted. To fix this, you’ll need to reflash your SD card so that you can start fresh.

Note: reflashing your SD card will delete any files you’ve created or changes you’ve made, such as your Voice Kit credentials, so you might have to repeat some setup steps.

Updating your SD card

Follow the appropriate instructions to reflash the SD card for either the Voice Kit, Vision Kit, or Maker Kit.

Learning more

I want to learn more about using the terminal and Linux commands

Check out these resources from our friends at the Raspberry Pi Foundation: Conquer the Command Line and Linux Commands.

I want to learn more about working with Raspberry Pi

Check out the Gettting started with Raspberry Pi guide for an introduction and tour.

Product lifecycle

Product line changes may bring products such as the Vision Kit and Voice Kit to the end of their life cycle. Once an End-Of-Life (EOL) notice is posted online, you can continue to purchase the product until the Last Time Buy Date or until product inventory is exhausted. Our goal is to assist you in making your final purchases of the product subject to EOL and to help you smoothly transition to new products.

Voice Kit EOL notice

  • EOL Notification Date: July 20, 2021
  • Last Time Buy (LTB) Date: December 31, 2021

Voice Kit EOL notice

  • EOL Notification Date: July 20, 2021
  • Last Time Buy (LTB) Date: December 31, 2021

Vision Kit help

Assembling the Vision Kit

The flex cable is too large to fit into the Raspberry Pi connector

There are two flex cables in the kit: one is long and one is short. The long cable connects connects the Vision Bonnet to the Raspberry Pi Camera, while the short cable connects the Raspberry Pi to the Vision Bonnet. See for a picture of the short flex cable for the Raspberry Pi board.

I need to replace the camera flex cable

You'll need the Raspberry Pi Zero camera cable.

Using the Vision Kit

I get a syntax error when running the Joy Detector demo

Your Raspberry Pi board may not be running the AIY software image (this error happens when using Python 2.7 instead of 3.4). You’ll need to update to the latest SD card image and try again.

My computer doesn’t detect the Vision Kit when using a USB cable

Double-check that you’ve connected the USB cable to the left-side USB port on your kit (the one labeled “Data”). Also, be sure that your kit is connected to a power supply via the other USB port — it will need more power than what your laptop can provide on its own.

Note: it’s not recommended that you use the USB connector cable to another computer as the sole source of power for your AIY kit. This may cause a brown-out or other issues. Learn how to connect to your Vision Kit

I get an error when I try to run a demo

First, check that you’ve typed the command correctly. Capitalization matters! If that doesn’t solve the problem, you might be in a different file directory, which means the command tried to find the demo in the current folder but it wasn’t there. Try typing the following command and pressing enter to navigate to the right directory:

cd ~/AIY-projects-python/src/examples/vision

Then try to run the demo again. If that still doesn’t work, you may have another demo that’s currently running, which will prevent you from starting a new demo. First check that you’ve stopped the Joy Detector. You’ll need to do this each time you reboot your kit.

If you are running another demo, try these steps to stop it: Open the terminal that has the demo running by clicking with your left mouse button on it, and once your terminal is in focus, press Ctrl-C.

My kit crashes on startup

Make sure you’re using a USB power supply that’s 5V 2.1A or higher. The Vision Bonnet requires additional current from the Raspberry Pi and using a power supply with less than 2.1A can cause instability.

After initially starting, the demos stop working

If the demos crash and none of the other solutions above solve your problem, open a terminal on the Raspberry Pi and type dmesg | grep 'Internal error'. If this prints any line at all, then your kit might be suffering from a hardware power problem (issue 346). Try running the following command:

echo "over_voltage=4" | sudo tee -a /boot/config.txt

Then reboot the kit.

The camera is blocking my terminal window

If you are connected directly to your Raspberry Pi via mouse, monitor, and keyboard, the camera window might block your terminal. That’s okay, as your terminal is still there in the background. Press Ctrl-C after pointing your camera at a few objects to stop the demo and close the camera window. Then you can scroll up in your terminal window to see what the camera identified. If you want to see the terminal and camera preview at the same time, you can connect your Raspberry Pi to WiFi and then connect to it from another computer via SSH. See steps 48-63 of the Voice Kit guide for more information.

Troubleshooting tips

Check the MCU firmware version and status

pi@raspberrypi:~ $ cat /sys/bus/i2c/devices/1-0051/status_message
0.5 is the newest and OK means its happy :D

Check the MCU Error code

pi@raspberrypi:~ $ cat /sys/bus/i2c/devices/1-0051/error_code

0xdeadbeef means operation is normal (no pins in a state they do not support, i2c bus has no errors the mcu knows about)

Check product name from overlay

pi@raspberrypi:~ $ cat /sys/firmware/devicetree/base/hat/product
AIY Vision Bonnet

Check product version

pi@raspberrypi:~ $ cat /sys/firmware/devicetree/base/hat/product_ver

If you have version 1.0 of the Vision Kit (check here), this command will output 0x0001.

Check serial number for you board

pi@raspberrypi:~ $ cat /sys/firmware/devicetree/base/hat/uuid

Check who built it

pi@raspberrypi:~ $ cat /sys/firmware/devicetree/base/hat/vendor
Google, LLC

Voice Kit help

Assembling the Voice Kit

My kit doesn’t match the assembly guide instructions

There are several versions of the Voice Kit and Vision Kit, each with different assembly instructions. Check which kit version you have by going here:

Booting the Voice Kit

My kit won’t turn on even though it’s plugged in and the green LED is lit

Your kit might be running a file system check, which happens after your kit shuts down abruptly. It takes around 5 minutes to complete, so try plugging in your kit and waiting.

If it still doesn’t boot, your SD card image may be corrupt, so try installing a new image.

My mouse and keyboard are connected, but they don’t seem to work

  • Check step 49 of the Voice Kit assembly and make sure your mouse and keyboard are plugged into the connector labeled “Data”
  • If you're using a USB hub, make sure it’s powered.
  • Make sure you are using a 2.1A USB power supply for your kit.
  • Finally, try rebooting the device while the keyboard and mouse are plugged in. To do that, unplug your power supply, wait a few seconds, then plug it back in. It may take a few minutes to boot up.

I want to configure my Voice Kit to activate with a button press

You can do this with the Voice Kit version 1.0 (and Raspberry Pi 3). Learn how

I want my Voice Kit to active using hotword detection (“OK Google…”)

You can do this with the Voice Kit version 1.0 (and Raspberry Pi 3). Learn how

The audio checkpoint script doesn’t work (“Failed to find the VoiceHAT soundcard…”)

There could be one or more issues that is causing the audio check to not find the sound card. Try the steps below:

  1. Check if the driver loaded by entering the following in your terminal window:
lsmod | grep googlevoicehat

If it’s not, trying using the terminal to manually load the driver:

sudo dtoverlay googlevoicehat-soundcard
  1. Check the list of detected audio devices: aplay -l
  2. Verify proper connections of the hardware (soldered pins, bent 40 pin headers, etc.)

If none of the steps above solve the issue, then open a terminal window and enter the command “dmesg”. Email us the output at

Using the Voice Kit

I want to use my Voice Kit with other Google devices (Home, Chromecast, etc)

At this time the AIY Voice Kit is not compatible with other Google devices.

When I try to start my demo the terminal responds “src/ bash:src/ No such file or directory”

The demo files have moved to src/examples/voice/. Try running it from that path.

Extending the Voice Kit

I want to use the Assistant Library/gRPC Service/Cloud Speech API in different languages

  • For the Assistant Library (, you need to use the Google Assistant app on Android or iOS. After you run the demo, your Voice Kit appears in the list of devices (in the app settings) where you can set your language preference.
  • For the Assistant gRPC service, you can also set the language preference in the Android or iOS app (see the previous item), or you can specify the language programatically—for an example, see the source code.
  • For the Cloud Speech API, you can set the language programatically—for an example, see the source code.

For details using the Google Assistant SDK, see the SDK documentation.

I want to make my own hotwords

Custom hotwords is not currently available with the Google Assistant SDK. However, you can try other voice recognition services like the Google Cloud Speech API.