Closed sondresorbye closed 6 years ago
Can you share the full error log that you're getting and your .asoundrc
configuration?
The .asounrc file:
pcm.!default {
type asym
capture.pcm "mic"
playback.pcm "speaker"
}
pcm.mic {
type plug
slave {
pcm "hw:1,0"
}
}
pcm.speaker {
type plug
slave {
pcm "hw:0,1"
}
}
Here are to links of pictures of the error I receive in the cloud console: (https://drive.google.com/open?id=0B92FVcqvLD_YaHNQbEZSMlJuZUk ), (https://drive.google.com/open?id=0B92FVcqvLD_YSGlpajl1U3djN3c )
I encountered this same problem and below is my .asoundrc file.
pcm.!default {
type asym
capture.pcm "mic"
playback.pcm "speaker"
}
pcm.mic {
type plug
slave {
pcm "hw:1,0"
}
}
pcm.speaker {
type plug
slave {
pcm "hw:0,1"
}
}
I also tried googlesamples-assistant-audiotest --record-time 10
and googlesamples-assistant-pushtotalk
. It performed well.
But google-assistant-demo
and googlesamples-assistant-hotword
didn't work.
BTW, pre-testing for mic/speaker are all good.
Same issue, the response code is 499 and the method is google.internal.speech.S3OnePlatform.Assistant
I also had the same issue, but for some reason it worked when I restarted.
EDIT: Is this happening only to people who have edited the .asoundrc
file?
No luck after restarting and I don't have a .asoundrc file.
Try this page, that Google made. If you don't have an .asoundrc
file, you can make one and put this code in it (courtesy of google):
pcm.!default {
type asym
capture.pcm "mic"
playback.pcm "speaker"
}
pcm.mic {
type plug
slave {
pcm "hw:<card number>,<device number>"
}
}
pcm.speaker {
type plug
slave {
pcm "hw:<card number>,<device number>"
}
}
EDIT: Also, edit the code in \<pointy brackets> with your own device and card numbers.
Thanks @jacksonsevendelta, still not working. I'm getting the following error [496:523:ERROR:mic_failure_detector.cc(120)] Channel 1 has failed. Powers -inf and -inf
now.
Just got the same issue. It started after the Google Assistant crashed and I tried to restart it. A reboot fixes this issue, though. (I'm also unable to kill it witch CTRL + C)
Yes! I know I'm hijacking, but why can't you kill it with CTRL-C? It should be implemented. I managed to do so by making a start-up script (you can find it here), but the timeframe to press it is small.
Also, @chris-gunawardena, I can't really help after that, sorry. What Pi do you have? It's recommended for the Pi 3, so Assistant may have problems with your device.
I've got this issue as well. google-assistant-demo
hangs with ON_START_FINISHED status and doesn't continue from there. Cloud console is showing 100% error google.internal.speech.S3OnePlatform.Assistant
Recording and playback is working just fine when I test using the "Configure and Test the Audio"-page. I'm using PI 3.
@vixriihi can you confirm that the hotword "Ok, Google" or "Hey, Google" are not recognized?
Also did you try to raise the microphone sensibility according to the troubleshooting instructions
@proppy Nice it's working. I thought that the status text would change to something else, but indeed it's now working just fine. Thanks
From #70, After activating alsa plug-ins dmix and dsnoop, works well on orange pi pc on both of hotword and demo sample code, but under these activating conditions, pushtotalk sample code is stuck.
@capitalfuse can you share your .asoundrc
with dsnoop
and dmix
enabled that you're trying with the pushtotalk sample?
@proppy .asoundrc in the case of
gRPC pushtotalk sample code ---------->
pcm.!default {
type asym
capture.pcm "mic"
playback.pcm "speaker"
}
pcm.mic {
type plug
slave {
pcm "hw:0,0"
}
}
pcm.speaker {
type plug
slave {
pcm "hw:0,0"
}
}
demo or hotword sample code, as I need to use vol control with remote, added softvol plugin. ---------->
pcm.!default {
type asym
playback.pcm "plug:softvol"
capture.pcm "plug:dsnooper"
}
pcm.dmixer {
type dmix
ipc_key 321456 #any unique value
ipc_key_add_uid true
slave {
pcm "hw:0,0"
rate 44100
}
bindings {
0 0
1 1
}
}
pcm.dsnooper {
type dsnoop
ipc_key 821456
slave {
pcm "hw:0,0"
channels 2
rate 44100
}
}
pcm.softvol {
type softvol
slave { pcm "dmixer" }
control {
name "Master"
card 0
}
}
ctl.softvol {
type hw
card 0
}
After deleting rate 44100 from dmix and dsnoop, works well in all sample commands. Finally fixed .asoundrc like the following,
pcm.!default {
type asym
playback.pcm "plug:softvol"
capture.pcm "plug:dsnooper"
}
pcm.dmixer {
type dmix
ipc_key 321456 #any unique value
ipc_key_add_uid true
slave {
pcm "hw:0,0"
}
bindings {
0 0
1 1
}
}
pcm.dsnooper {
type dsnoop
ipc_key 821456
slave {
pcm "hw:0,0"
channels 2
}
bindings { # for output mono sound to 2ch right and left speaker, need this item.
0 0
1 1
}
}
pcm.softvol {
type softvol
slave { pcm "dmixer" }
control {
name "Master"
card 0
}
}
ctl.softvol {
type hw
card 0
}
ctl.!default {
type hw
card 0
}
I made it work by buying a new mic. My previous one was very old.
fixed pcm.dsnooper in .asoundrc. please see the above.
pcm.dsnooper {
type dsnoop
ipc_key 821456
slave {
pcm "hw:0,0"
channels 2
}
bindings { # for output mono sound to 2ch right and left speaker, need this item.
0 0
1 1
}
Chris - how did you solve your issue?
I'm having the same problem:
(env) pi@raspberrypi:~ $ googlesamples-assistant-hotword ON_MUTED_CHANGED: {'is_muted': False} ON_START_FINISHED [3128:3145:ERROR:mic_failure_detector.cc(120)] Channel 1 has failed. Powers -inf and -inf
@vixriihi Can you please explain what helped you to fix it? or what steps you specifically followed to fix it?
I had google assistant sdk working on raspberry pi 3 by following the instructions provided here . I tested it few times and it was working fine. I'm using new MIC and speaker. I turned off the Pi for few days and wanted to demo it again but I started running into the below issues
(env) pi@raspberrypi:~ $ google-assistant-demo
ON_MUTED_CHANGED: {'is_muted': False} ON_START_FINISHED
I am also getting an error in the google cloud console: Method google.internal.speech.S3OnePlatform.Assistant
Please provide some inputs to help me fix this issue.
Just shutdown Pi and turned on again, fixed it. I wonder what changed. Please provide some information if anyone of you know what actually happened.
I'm a bit confused about what actually fixed it.
Can you try with the latest version of the library?
google-assistant-library 0.1.0
I'm getting the same error. I've tried the Assistant SDK (Python), the AIY Projects Raspbian repository, as well as the AIY Voice Kit image and instructions. Nothing works. After two days and a lot of frustration, I haven't managed to get a response from "Okay Google."
The AIY Projects page estimates you'll be up and running in an hour and a half, while the Assistant SDK calls it a "turnkey solution," and the Rasbian Project promises an "easy-to-use API." I'm sorry, but no.
Since the initial issue, we've updated the Python library multiple times. Any bugs with audio configurations may have been resolved. If you are still having issues, please make a new issue.
Hello, have same problem. Mic test is good, but google can't recognize me. -audiotest result is confusing. .asoundrc is here:
pcm.!default { type asym capture.pcm "mic" playback.pcm "speaker" } pcm.mic { type plug slave { pcm "hw:1,0" } } pcm.speaker { type plug slave { pcm "hw:0,0" } }
I have installed the Google assistant on my raspberry pi 3.I have followed the documentation for the google assistant sdk, but it does not work. The assistant is not responding when I say "Ok, Google"
The output when I try to start the assistant is:
When I test the recording, the recording is being played too fast. Can this be the problem that the Google assistant is not responsing?
I am also getting an error in the google cloud console: Method google.internal.speech.S3OnePlatform.Assistant