How to Setup an Orange Pi Zero DIY Smart Speaker with Google Assistant SDK

Orange Pi Development Boards

A preview release of Google Assistant SDK working with Raspberry Pi 3 and other ARMv7 boards was released in May, and soon after, AIY Projects Voice Kit was offered for free with Raspberry Pi Magazine in order to a complete smart speaker kit working with RPi 3. I wanted to try it on one of FriendlyELEC or Shenzhen Xunlong Allwinner board, since all we need is audio input and output, and an Internet connection. Earlier this month, I came across Orange Pi Zero Set 6 Kit that had all I needed: Orange Pi Zero ARM Linux board, an expansion board with built-in microphone and audio output jack, and a cute and small case to neatly put everything together.

Orange Pi Zero Set 6 Kit Unboxing and Assembly

Shenzhen Xunlong sent me the kit so that I can try it out.

The package includes two Orange Pi packages, the plastic case, some a bag with screws and rubber pads.

One of the package comes with Orange Pi Zero board powered by Allwinner H2+ quad core processor with 512 MB RAM, micro USB for power, 10/100M Ethernet, a USB port, a WiFi antenna, and expansion headers on the top…

Click to Enlarge

.. and a micro SD card slot on the bottom, which we’ll use to boot the operating system.

The second package is for Orange Pi Zero Interface board v1.1 with two more USB port, an IR receiver, a 3.5mm audio jack, and a built-in microphone (top right).

Click to Enlarge

The next step is just to plug the interface board into Orange Pi Zero board, bending the WiFi antenna between the two boards…

Click to Enlarge

… before pushing the board from the bottom of the enclosure, tightened the cover with the 4 screws, and adding the 4 rubber pads.

Click to Enlarge

Installing Ubuntu on Orange Pi Zero & Configuring Networking and Audio Input & Output

Google Assistant SDK requires Ubuntu or Debian operating system, so I downloaded Ubuntu server legacy image on Armbian website, since the mainline image does not support XR819 WiFi module at all. I then extracted Armbian_5.30_Orangepizero_Ubuntu_xenial_default_3.4.113.7z and flashed Armbian_5.30_Orangepizero_Ubuntu_xenial_default_3.4.113.img  with Etcher on a micro SD card using my main computer.

Once it was done, I took the micro SD card to Orange Pi Zero board, connected my own speakers to the 3.5mm audio jack and USB for power, as well as a 5V/2A USB charger to power the system.

Click to Enlarge

You may notice an extra USB dongle on the picture above, but we’ll talk about that later.

I’ll use WiFi to test Google Assistant, but I also connected an Ethernet cable to make it easier for first time setup. If you don’t have Ethernet, you could also connect the board via a USB to TTL board to access the serial console.

After looking up the IP address of the board in my router, I connected with SSH using root (password: 1234):

The very first boot it will ask your to change the root password, and create a new user. You may want to create “pi” user at this stage. We should not be able to access the command, and run the following command to configure Orange Pi Zero board:

I first changed the timezone.

Then selected WiFi – Connect to wireless access point to connect to my WiFi router.

We can quit armbian-config, and optionally disconnect the Ethernet cable, and reconnect SSH over WiFi, which I did.

Back in the command line, I tested audio recording & playback with the built-in microphone and speakers using the same commands as in ReSpeaker guide:

The first command recorded my voice, and then I pressed Ctrl+C to stop, and play it back with the second command. I worked fine for me. If you wish you can adjust the playback and recording volume with:

We’ll also need to create .asoundrc file for Google Assistant to work with ALSA. For we need to note the microphone card and device number (0,0):

as well as speaker card and device number (0,0):

If you want to use your TV as speaker, you’d use Card 1, device: 0.

If you have not created a pi or other user yet, you’ll want to add one belonging to sudoers and audio groups, and go the home directory:

Now create a ~/.asoundrc with the following match the card and device for our mic and speaker:

[Update: As we’ll see below this won’t work with the built-in microphone, but as indicated in the comments below, this can be fixed by changing pcm.mic section with:

I have not changed the rest of the post, but the built-in microphone on Orange Pi Zero does work now]

Setting Up Google Assistant on Orange Pi Zero

Now that audio and networking are both setup and working, we can carry on with the instructions to install Google Assistant SDK and demo. Those are the same for all boards, and I’ll describe them below in details.

We need to configure a Google Developer Project.

Go to the Project page in Google Platform Console, and click on Create Project.

I called it Orange Pi Zero Smart Speaker, and clicked on Create.

Next, we’ll need to enable Google Assistant API for our project. Simply click on Enable on that page. The “Orange Pi Zero Smart Speaker” project was already selected in my case.

We’re then being asked to create an OAuth Client ID. I only filled my email address, and a product name before clicking on Save.

Select Other in the next page, and give another name – I used “Linux Thingy” – before clicking on Create.

You should now see an OAuth client pop-up window with your client ID and secret. You can just click OK, no need to copy or save either.

Instead we’ll download client_secret_[your_client_id].json file on the next page by clicking on the arrow circled in red.

I then copied that file to Orange Pi Zero board using scp (change [your_client_id] with your actual client ID):

Now that we have configured most options we need in Google Cloud platform, let’s install Python 3 and dependencies:

At this stage to need to make sure “Web & App Activity” with “Include Chrome browsing history and activity from websites and apps that use Google services”  checked, Device Information and Voice & Audio Activity are enabled in the Activity Controls page. All options were already enabled for me by default.

We can now install Google Assistant SDK package including library and sample code:

and the authorization tool:

Run the authorization tool:

This should show a text reading “Please visit this URL to authorize this application: https://….”. Select the URL and open it in a web browser, and copy/paste the authorization code you see in the web browser back into the terminal.

We can now start the demo

That’s the initial output

I went to say “OK Google” and “Hey Google”, and… nothing. No answer, and no info in the terminal window either. I tried several times, and also played with:

as explained in the troubleshooting section in Google Assistant documentation, incrementing the value by 6 until I reached -60, before running the demo, and no luck. So I thought maybe my “American accent” could have been the problem, so I launched Google Translate in my phone, and make it say “OK Google”. Same disappointing result. I went to Google Cloud Console, and good see some activity, but with 100% errors.

Click to Enlarge

So I did some searching on the web, I could find one person changing the .asoundrc file, but I did not work for me with the file generating plenty of sound related errors when running the demo, and another fix a similar problem by changing the microphone. Which reminded me, I had bought a cheap USB microphone on eBay for $1.45. So I connected it to Orange Pi Zero board, and that’s the extra USB dongle in the photo with the speakers higher in this guide, and check the card and device number:

We can see the USB card 2, device 0, so I update the relevant line in pcm.mic section in .asoundrc:

and ran the demo again. Success! The device would reply to make “OK Google” request, and tell me the time, my name, the weather for tomorrow, sing a song, do some calculus and so on.

This is what the output looks like when the demo works:

Bear in mind that Google Assistant SDK is in preview mode right now, and it’s mainly aimed at developers. First, I understand you can’t do automation tasks, or play music like you can on an actual Google Home. The good news is that you can roll you own implementation, and for example, somebody made his AIY Project Voice Kit work with Google Music, something you should be able to reproduce on other boards. Google Home also includes a microphone array, which will make hot word detection more reliable even in noisy environment. Orange Pi Zero + USB microphone did a decent job though, as I could use it 5 or more meters away.

Next time you boot the board, you can login as pi user, and run the following commands to run the demo:

You could also use systemd to automatically start the demo, or your own implementation, and I did so with instructions adapted from Nordic Semi’s Google Assistant Wiki.

First, create /etc/systemd/system/google-assistant-demo.service file with the following content:

I add “Type = idle” as other the app would not run, possibly because it would start after the audio was enable. I did not find any audio specific services handled by systemd, so “type = ilde” will make sure the service starts after other are loaded.

Now we can enable the service, and start it:

The demo will run at this point, and it will start automatically each time your boot the board. I played with it for several hours and it seemed stable. Some people have however expressed concered about WiFi on Orange Pi Zero, so if you encountered some instability issues, you may want to switch to Ethernet instead, or use a WiFi USB dongle, or use another Orange Pi or NanoPi board with a time-proven WiFi module.

I’d like to thank Shenzhen Xunlong Software for sending the kit my way. If you are interested, you can purchase it for $12.95 + shipping.

102 Replies to “How to Setup an Orange Pi Zero DIY Smart Speaker with Google Assistant SDK”

  1. Array mics are the defining piece of the smart speaker category since they allow it to work from more than a couple of feet away. Does the Google SDK support multiple mic inputs?

    Array mics work by detecting the hot word in the input stream. Once the hot word is located the phase difference between the multiple copies of the hot word is measured (one copy per mic). Then do some math magic and all input that does not have that same phase delay is removed from the input stream. So you could have a TV sitting off at a different angle and the mic array will remove it from the audio input and only focus on the person speaking. With a single mic there is no noise removal.

    ESP32 has ability to have 2 mics. With some code it can be turned into a cheap array mic chip. Dedicated array mic chips are about $7.

  2. Does this work on a Orange Pi Lite?
    Can i take risk of ordering 32GB Samsung Evo card or do i need to have it smaller?

  3. @Jon Smirl
    Don’t know about multiple mic support.
    “OK Google” works even 10 meters away with that setup. Maybe the microphone being outside the box helps too. I think it picks up 95% of my “OK Google” requests.

    I did not have a TV playing at the same time, or much of other audio at the same time, as you explain that’s when the noise cancellation helps the most.

  4. @Fossxplorer
    All you need is audio input and output, and network connectivity.
    With Orange Pi Lite, you’d still need a USB sound card with output for your speaker, and maybe including a microphone too, since I’m not sure the mic on Orange Pi Lite will work, as I had troubles with the one in Orange Pi Zero.

    I’m using a 32GB micro SD card in my setup. You can get something smaller if you want too.

  5. I followed the instructions on your previous blog. And modified to give it a response sound. It works great.

    The Wifi on the opi zero is my biggest problem. It’s often unreliable, google simply tells me ‘there was a glitch, please try again’. And i didn’t get the expansion board so I’m reserving the usb port for a Samson Go Mic. My only option is ethernet which IMO is not very usable. I may have to look into that expansion board for usb ethernet.

  6. Try to insert “format S16_LE” within mic.slave section in ~/.asoundrc file. This solved “OK Google” recognition problem in my case.

  7. @Squalazzo
    Memory usage on mine with the demo running:

    It should be no problem with 256MB RAM.

  8. @Kaiyoti
    WiFi is pretty stable on mine. I can use it all day.
    However, yesterday I lost ssh access at one point, and when pinging the host seems down. But I was still able to use Google Assistant. Very odd problem. A reboot fixed that.

  9. Whew, seems like a lot of work though it would be cool if this was available to people who didn’t know coding and Linux well. I’ll just buy the Google home.

  10. @cnxsoft
    Wrt Wi-Fi on OPi Zero according to Xunlong PCB rev 1.4 (that’s the one you have) should ‘fix’ Wi-Fi issues. At least they changed something power related (a web search for ‘New OPi Zero – Yet another high temperature issue…’ should be sufficient, in post 3 there are some close-ups of PCB differences). PCB rev 1.4 also has for whatever reasons higher internal temperature readouts — more details also at the end of the aforementioned thread.

    The other issue you ran into — no SSH access through Wi-Fi after some time — seems to be related to network manager. At least we had several reports affecting various boards reporting the same and establishing Wi-Fi the old way with wpa_supplicant seems to fix it. Some users reported they would be only locked out from Windows boxes while with Linux they would still be able to SSH in. I was never able to reproduce (all my productive hosts running macOS and I didn’t extensively test). Also issue seemed to be gone with mainline kernel –> Bottom of ‘Orange Pi Lite SSH Problems’ thread in Armbian forum.

  11. @tkaiser

    A lot of the issues with the xr819 seem to be broken frames so if they fixed the antenna or power supplies for it that might actually improve it. The driver for the chip is still a bag of poo though. 😉

  12. Also you can install and test alexa through This software can trigger following voice command by “alexa” voice (using by pocketsphinx) or GPIO button. Then, it needs delayed voice input, nevertheless having not good recognized result, i guess, partly by worse performance compared to RPi 3.

  13. dgp :
    xr819 … The driver for the chip is still a bag of poo though.

    Of course. But what puzzles me is: people talk since months about OPI Zero PCB rev 1.4 overheating (which is mostly wrong temperature readout and not really overheating… which is a real problem due to throttling jumping in too early based on the temperatures being reported way too high). Xunlong’s explanation that this is due to ‘fixed Wi-Fi’ is also known since at least a month. And literally no one having an OPi Zero 1.4 in his hands is going to test this. It’s so f*cking easy: put your OPi Zero 2 m away from the AP and use ‘iperf3 -s’ on a wired host connected to the same AP/router, use ‘iperf3 -c’ and ‘iperf3 -c -R’ to test TCP/IP bandwidth, done. Takes 5 minutes max but no one spends this amount of time. I already though about ordering another OPi Zero but it feels somewhat stupid trying this on my own since everyone can contribute that easy.

  14. aquilegia :
    partly by worse performance compared to RPi 3.

    Hmm… I just looked into my archives. With code that makes use of NEON I see RPi 3 being approx. three times faster than an H3/H2+ device running at the same clockspeed with same amount of CPU cores active. When talking about usual integer performance RPi 3 has an advantage of maybe 130%-140% compared to H3/H2+ devices running also at 1.2GHz. Applications that depend on memory bandwidth/latency are more affected since RPi 3 memory is a lot faster compared to H2+/H3 devices with single bank DRAM configuration (especially when using Armbian since we chose by default to use ‘IoT settings’ on those small H3 devices focusing on low consumption and underclock DRAM massively).

    Which OS image are you currently using on your Zero? Just asking since still some vendor OS images for H2+/H3 devices use Allwinner’s crappy thermal/throttling settings leading to killed CPU cores once throttling occured the first time. In case you’re not using Armbian better check from time to time with

  15. @Athar
    NEON is neither related to HW accelerated video decoding nor to shitty Allwinner default thermal/throttling settings (that’s all what the above article in Armbian forum was about).

    Why not looking up first what NEON is (a web search for ‘Advanced SIMD (NEON)’ should be sufficient)?

    NEON code on ARMv8 platforms (like RPi 3) can run much faster than on older ARM architectures (like ARMv7 on an OPi Zero). Since @aquilegia talked about ‘worse performance’ maybe he already looked into code and has an idea whether these ‘spy on people in their home’ APIs from Google or Alexa make use of NEON or not.

    If we’re talking about NEON then ARMv6/ARMv7 CPU cores have a clear disadvantage and moving from an OPi Zero or NanoPi NEO to an ‘OPi Zero Plus 2 H5’ or ‘NanoPi NEO 2′ or NanoPi NEO Plus 2’ (how I hate these stupid names!) could solve performance issues simply by switching from ARMv7 (H2+/H3/BCM2386) to ARMv8 (H5/BCM2387).

    1. @tkaiser

      Ok thanks. I was always confused about V7 vs V8 code advantage.

      So V8 is far superior, and CHIP A13/R8 is V8?

  16. @Jon Smirl

    @Jon Smirl, You said, “Dedicated array mic chips are about $7”.

    Can you post a link please? I’ve seen cheap single I2S MEMS mics on a chip before, but never a mic array on a chip, much less one for $7.

    Here is a small mic array board with an XMOS commutating multi-core micro-controller capable of doing parallel DSP. But it’s not an array on a chip, and it certainly costs more than $7 (I think it goes for around $80-$100).

    Another device using the same approach is the $79 ReSpeaker by Seeed Studio.

    Note, I’ve used the XMOS xCore chips that are used on these array mic boards before, but for other applications. The XMOS chips are nice once you get familiar with programming a commutating parallel processor, but the bad XMOS documentation and user-hostile implementation of their Eclipse-based IDE made working with the XMOS parts painful (in my opinion). BTW, if I remember correctly XMOS is based in the UK.

    Thanks & Regards, David

  17. @tkaiser
    In my setup (About 3 meters away + thin wall):

  18. @cnxsoft
    and Tx direction:

  19. @cnxsoft
    Thank you for the numbers. Given that for whatever reason every second wireless frame gets dropped/retransmitted the performance would be on par with the other usual suspects when we’re talking about 1T1R/2.4GHz Wi-Fi once the driver could be fixed (which would require Allwinner providing real documentation which most likely will never happen).

  20. Hi.
    when i try to run the demo, i get the following error:
    [] Input error

    any advice?


  21. Hello.
    I’m trying to scp my client ID into my Orange Pi Zero, but it returns with ‘No such file or directory’,
    even though I’ve downloaded the client_secret_[your_client_id].json file onto my computer.
    Any suggestions? Thanks.

  22. @OrangePro
    That just means one of the path in the command line is incorrect, either the source file in your host PC, or the destination directory in Orange Pi Zero.

    Another possibly is some special characters in your_ client_id in which case you need to escape the file name

  23. @OrangePro
    So you’ve already copied the file to Orange Pi Zero successfully, and it’s just not finding the file when you run the command with google-oauthlib-tool? It should just be client_secret_xxxxx.json path/filename is wrong then.

  24. To get the beep back you can change the file in /env/lib/python3.5/site-packages/google/assistant/

    i added the following lines

    Actually it can be anything you want. I have the original beep from my phone.

  25. @cnxsoft
    So… this is what I’m typing into PuTTY;

    scp [email protected]:~

    ^ < this is what the downloaded .json file is called, which is in my Downloads folder.

    And this is what it results in;

    [email protected]'s password: No such file or directory
    [email protected]:~#

    Thanks for any step by step help to this problem.

  26. @Orange
    Are you running that command in Orange Pi Zero?
    You need to run it in the computer where you downloaded the file. If you are using Windows, then using Windows Susb-system for Linux to type the command, or maybe simpler, WinSCP.

  27. @cnxsoft
    I finally got the Google Assistant working thanks to your WinSCP suggestion. I really want to be able to use Google Assistant, as soon as the power adapter is plugged into the Orange Pi Zero, while connected to the internet via an ethernet cable.

    I followed the last set of instructions, and was able to use Google Assistant without typing source env/bin/activate and google-assistant-demo, while on the client program. I then tested to see whether I could forgo the client program, but got no reply. So I went back on the client program to see whether it still worked without source env/bin/activate and google-assistant-demo, but got no reply.

    So, I thought I’d re-type sudo systemctl enable google-assistant-demo.service and sudo systemctl start google-assistant-demo.service on the client program, but it now results in this:

    Failed to start google-assistant-demo.service: Unit google-assistant-demo.service is not loaded properly: Invalid argument. See system logs and ‘systemctl status google-assistant-demo.service’ for details.

    So, I thought I’d delete the /etc/systemd/system/google-assistant-demo.service file on WinSCP and re-do everything again, before I asked you guys for help, but I’m not able to do this because:

    General failure (server should provide error description).
    Error code: 4
    Error message from server: Failure

    Common reasons for the Error code 4 are:
    – Renaming a file to a name of already existing file.
    – Creating a directory that already exists.
    – Moving a remote file to a different filesystem (HDD).
    – Uploading a file to a full filesystem (HDD).
    – Exceeding a user disk quota.

    I also tried removing it via command, but can’t because:

    /root$ rm /etc/systemd/system/google-assistant-demo.service
    rm: cannot remove ‘/etc/systemd/system/google-assistant-demo.service’: Read-only file system

    How do I change access permission to the file, so that I can delete it (chmod 7, chmod 777 doesn’t do anything)?

    Thanks for any help.

  28. @Orange
    Better only use WinSCP for the transfer of the file from your PC, and use the terminal for everything.
    As for you need error it looks like the file system may be mounted read-only due to corruption.

    mount command with no option should show the list of mounts with the mount status (ro = read-only, rw = readwrite).

  29. I want to include led like voicekit but I get this error
    I use “”
    [email protected]:/home/pi# source env/bin/activate
    (env) [email protected]:/home/pi# google-assistant-demo
    Traceback (most recent call last):
    File “/home/pi/env/bin/google-assistant-demo”, line 7, in
    from google.assistant.__main__ import main
    File “/home/pi/env/lib/python3.5/site-packages/google/assistant/”, line 22, in
    from pyA20.gpio import gpio
    ImportError: No module named ‘pyA20’

  30. First, thanks for the detailed instruction! Based on that I was able to install and run google assistant. However I run into a problem: As soon as I plug in my speaker into the AV out there is a constant and loud static noise in one of the two speakers. Changing speakers do not help, and I got the same noise even if I just plug in the headphone. The other speaker has no noise at all and gives the proper sound (in fact the noisy speaker produces no usable audio output).

    I tried several power cables and power supply to no avail. I even tried to power the unit with a battery powerbank and I still got the static noise.

    Any suggestion on how to fix this, or is this a hardware problem? Or maybe it is mono only? I’m afraid that it would cost more to return my unit then to get a new one 🙁

  31. @saon
    That one is standard too, but the number of “rings” may differ between different plugs.
    If you have pushed it all the way, try to pull it out a little and see what happens. Maybe play music through the speaker, and adjust the plug position step by step to check if both speakers output audio.

  32. @cnxsoft

    Thanks for your suggestion. Turns out the circle board was not sitting properly inside the case — not by much, but just enough that the AV jack could not be inserted all the way. Following your suggestion, I resembled the whole thing and it now works. THANKS!

    Another unrelated question: I’d like to put the OP speaker in my office where I have access to wifi but not the router. How would I enter the wifi connection’s name and password?

  33. @SaOn
    Several solutions to your last point:
    1. Connect a USB to TLL debug to access the serial console in your location
    2. Connect the board to a TV via the AV port
    3. Setup an access point with armbian-config, then use your phone or PC in your office to connect to the board, and connect to your router
    4. Manually edit to configuration files before moving to office (not sure where that is yet)

  34. Excellent work. Fortunately, I didn’t face the mic issue and is working fine for me from a maximum distance of 5 feet. Beyond that it does not work. Thanks for sharing the detailed instructions. Next for me is:
    1. Connect orange pi to a bluetooth speaker
    2. Connect to WIFI using /etc/network/interfaces and /etc/wpa_supplicant/wpa_supplicant.conf. This is something, I am still not able to achieve. I get error when I try to do this. Finally, I used armbian-config as mentioned in your tutorials.

    ERROR: run-parts: /etc/network/if-pre-up.d/wpasupplicant exited with return code 1

    Following is what I am doing =================OrangePi Zero and Armbian Ubuntu
    Open Wifi Configuration file
    1. sudo nano /etc/network/interfaces
    #Paste the following lines
    auto lo
    iface lo inet loopback

    auto eth0
    allow-hotplug eth0
    iface eth0 inet dhcp

    allow-hotplug wlan0
    iface wlan0 inet manual
    wpa-roam /etc/wpa_supplicant/wpa_supplicant.conf
    iface default inet dhcp

    #Open WIFI Config File
    2. sudo nano /etc/wpa_supplicant/wpa_supplicant.conf

    # Copy and paste the following and set your SSID and PASSWORD

    #Turn Wifi ON
    3. sudo ifup wlan0

    #Update Time Zone
    4. sudo dpkg-reconfigure tzdata


  35. Hello, thank you very much for the tutorial, I worked at the first shot!
    I am now wondering, is there any way to change the lang (to another one supported by Google Home/Google Assistant app)? I have searched a bit, without success 🙁

    1. It tells me the same thing when I ask…
      I noticed that, but this seems to be only for custom actions, switching languages might be implemented in next releases (I hope so)… I also searched for a way to change regional settings (units, date formats) without more success.
      Thanks for your reply!

  36. Hi there and thank you for this tutorial.

    Unfortunately, I am stuck.
    At “source env/bin/activate” it tells me “Permission denied” with sudo prefix it tells me “command not found”.
    If i run this command as root, instead of the pi user, it works, but then again, in the next step “python -m pip install –upgrade google-assistant-library” it gives me this error: “no module named pip”.

    Do you have any advice for me, how to make this work? Thank you kindly.

  37. @cnxsoft
    Thank you for your reply. But I do not know how to check the content of it? nano? Or how to change the permissions. Is it the CHMOD Command?
    Thank you for your time.

  38. if i paste the code in the terminal will it automatically create a .asoundrc file? and do i need to install winscp to copy the file in orange pi zero or just pasting the code above will work?

  39. @Hanzala
    You’d need to use a tool to create a file. You may want to use “nano” in the command line:

    Then copy/paste the code, press “Ctrl” and x keys, and select yes to save.

    Good luck.

  40. I’ve followed this twice now on two Orange Pi Zero’s and I am really pleased to say it works perfectly – I am using the built in mic and added a non-powered speaker. All good. There have been, I think, slight changes in the way Google shows the web pages when you are configuring the API, but not too tricky to work out!

    However, one of my two Orange Pi Zero’s runs very hot when running this – so hot in fact that the case is barely touchable. It doesn’t seem to affect responsiveness of the “Ok Google” prompt but I am reluctant to leave it plugged in less it melts. The other one, gets warm but not overly.

    What could be wrong – the two were bought about 6 months apart, and it is the newer one that gets roasting hot. It can’t be a configuration thing can it? The only difference is their proximity to the WiFi router – the cool one is only a few feet away, the hot one is one floor and several walls away. Could that have a bearing? I can see the far away one does time out on PING every now and then.


  41. @Bungalow Bill
    Most probably on the hot OPi Zero you can read ‘Rev 1.4’ while the colder one is an older PCB revision? Even if that’s not a software/settings issue there’s more than one thread discussing these temperature problems in Armbian forum so that would be the first place to look…

  42. everything works fine but when i try to install google assistant library using the code above it shows lots of erros in red colour and the library does not it because of my internet connection or something else?plz help

  43. (env) [email protected]:~$ google-assistant-demo
    usage: google-assistant-demo [-h] [–credentials OAUTH2_CREDENTIALS_FILE]
    google-assistant-demo: error: argument –credentials: No such file or directory: “/home/pi/.config/google-oauthlib-tool/credentials.json”

    i am getting this error.what do i do????plz help

  44. Hi @cnxsoft
    I’m trying to do this project on a Orange Pi PC, google assistant recognizes my voice pretty good but apparently there is a problem with my speakers, I keep getting ‘ ALSA lib pcm.c:7963:(snd_pcm_recover) underrun occurred ‘
    Any idea how to solve this?

  45. (env) [email protected]:~$ python -m pip install –upgrade google-assistant-library
    Collecting google-assistant-library
    Could not find a version that satisfies the requirement google-assistant-library (from versions: )
    No matching distribution found for google-assistant-library
    (env) [email protected]:

    I am getting above error. Not able to install google-assistant-library
    I am using orange pi zero plus 2 H5 with following config
    Welcome to ARMBIAN 5.27.170614 nightly Ubuntu 16.04.2 LTS 4.11.1-sun50iw2

  46. Hello to all, i hope you are doing well.
    Great tutorial, good job 😉
    I lunch my assistant and work perfectly fine on cubieboard A10 a.k.a (1) with Armbian_5.31_Cubieboard_Debian_jessie_next_4.11.6 🙂
    Now i want when i say “OK Google” or “Hey Google” some indication whether she hears, led or sound … something that let me know she is ready for me 😀
    Try this :

    August 17th, 2017 at 06:11 | #33
    Reply | Quote

    To get the beep back you can change the file in /env/lib/python3.5/site-packages/google/assistant/

    i added the following lines

    import os

    if event.type == EventType.ON_CONVERSATION_TURN_STARTED:
    os.system(‘/usr/bin/aplay google_now_voice.wav &’)

    Actually it can be anything you want. I have the original beep from my phone.

    But everything burn down to hell ;s

    How i can accomplish that ?

  47. cnxsoft :
    The real Google Home can be used with IFTTT. But the Google Assistant SDK not. You’d have to dig into the source code and your device API to enable this.

    The Google Assistant SDK can be used with IFTTT, too

    but I guess I can just use the text transcript by adding something on /env/lib/python3.5/site-packages/google/assistant/

    Get the transcript of the user request

    The Google Assistant SDK gives you a text transcript of the user request. Use this to provide feedback to the user by rendering the text to a display, or even for something more creative such as performing some local actions on the device.

    The transcript is located in a Python string object in the ON_RECOGNIZING_SPEECH_FINISHED event.



    The Assistant has determined the final recognized speech.

    text(str): The final text interpretation of a user’s query.


  48. [email protected]:~$ sudo systemctl enable google-assistant-demo.service
    Failed to execute operation: No such file or directory
    [email protected]:~$ sudo systemctl start google-assistant-demo.service
    Failed to start google-assistant-demo.service: Unit google-assistant-demo.service not found.

    i am getting this error

  49. @cnxsoft
    Hi cnxsoft,
    Can you please explain in details. I cannot make the sound feedback working.
    Where and how can I call aplay. What is the “right location” ? And how can I know the full path of a file ?
    Thank you !

  50. @Yuyu
    Sorry, I can’t, as I’ve not looked into it in details.
    You need to learn about “Linux command line” and “Python programming basics”, and then you’ll be equipped to solve those very common issues.

  51. Hi cnxsoft,

    Thanks for your fast reply.
    My problem was having the assistant running automatically from reboot. When I disabled this option and ran the assistant manually, the .wav worked.

  52. By the way, do you have any idea how to automatically run google assistant at each reboot, but still being able to open wav files ?

  53. Thanks for your post!

    But what about built-in microphone? Can anybody use it with built-in microphone?
    What is the content of .asoundrc in this way?

  54. i cant really get the client_secret to work, i always get an error message, it was posted a couple of times in the comments already but i cant figure it out. I downloaded the file and what do i do now? (ive downloaded winSCP aswell but dont know where to put the file). It would be nice if you guys helped me out a bit.

  55. installed latest armbian 5.35 xenial on orange pi zero, i was going to setup this, but something seems missing… i’ve the addon board, with mic and 3.5 jack output
    [email protected]:~# arecord -l
    arecord: device_list:268: no soundcards found…
    [email protected]:~# lspci
    pcilib: Cannot open /proc/bus/pci
    lspci: Cannot find any working access method.
    [email protected]:~# alsamixer
    cannot open mixer: No such file or directory

  56. worked great on my orangepi useing Armbian_5.35_Orangepizero_Ubuntu_xenial_default_3.4.113 .just followed your guide and bingo .. cheers

  57. Great tutorial, thank you!

    Almost at the end I’m running into a problem though: when issuing google-assistant demo I get this

    “the following arguments are required: –device-model-id”

    I have tried to find out how to specify this but in doing so I encounter another problem: the command google-assistant-devicetool gives me “command not found” and that is the one, if I get understand it correctly, I should use to set device model.

    I don’t see device model anywhere in your tutorial so I’m a bit mystified – what am I missing?
    Answers, suggestions, ideas are all very welcome 🙂


  58. @Jorg Muskens Where abouts in the script would this be added?

    @Dag Hovden
    Add it to the end of the start command, i.e:

    From what I can ascertain, it doesn’t matter what you call it.

  59. @cnxsoft
    Hello a newbie question
    I intent to make and use this for home automation(control a few sonoff or something similar) can I use google home app with this? what happens if I cut the power? Will I have to SSH into it and start it again or is just plug and play?

  60. @Akki
    The app will automatically start at boot time, so no need to SSH to it to restart the app after a power failure.
    The previous SDK did not work with Google Home app, but it might be different with the December release which I have not tried yet.

Leave a Reply

Your email address will not be published. Required fields are marked *