Comments (12)
ensonic
commented on April 28, 2024
1
This article shows how to modify the alsa config:
http://www.androidauthority.com/build-google-assistant-raspberry-pi-770296/
from aiyprojects-raspbian.
devilmind
commented on April 28, 2024
1
I have solved myself using rapi-config,
If I force to use the jack option the .asoundrc file will not modified on reboot and the config remains in place.
from aiyprojects-raspbian.
drigz
commented on April 28, 2024
The voice recognizer should work with any audio setup if ALSA is configured correctly. The ALSA configuration can be difficult, though, and depends on your hardware setup (eg USB soundcard, onboard audio, HDMI audio, bluetooth speaker). We don't plan to provide instructions for setups other than the AIY Projects Voice Kit.
#7 has some more info. Maybe you could post your ALSA config here if you get it to work, so others can use it as a guide?
from aiyprojects-raspbian.
dylandignan
commented on April 28, 2024
@drigz thanks for following up. I have the setup working when running it in the "dev console" but when it runs as a daemon I am getting errors using the exact same arecord and aplay commands that the voice recognizer is using. I simply add the bluetooth speaker/usb mic as the default Pulseaudio sink/source and the dev console works as well as the check_audio script.
Is there anything that the daemon does differently which would cause this? I will do some more digging tonight.
from aiyprojects-raspbian.
drigz
commented on April 28, 2024
Ah, I see. You can inspect systemd/voice-recognizer.service to see how it's started. A couple of possible differences:
- Are you running as user
pifrom the terminal? - Are you starting the service with
sudo systemctl start voice-recognizeror by enabling it and restarting the Pi? - Are you passing the
-I/--input-deviceor-O/--output-deviceparameters which wouldn't be used by the service? - When it starts as a service, what errors do you see with
sudo journalctl -n 20 -u voice-recognizer?
Also, the voice-recognizer doesn't use PulseAudio, so you need to make sure the default ALSA device is correct. However, if check_audio is working, then it should be correct.
from aiyprojects-raspbian.
salmanfarisvp
commented on April 28, 2024
@diggy106 Can you please Explain how to setup Speaker and USB mic with this AIY Image .
Thanks in Advance
from aiyprojects-raspbian.
dylandignan
commented on April 28, 2024
Are you running as user pi from the terminal? - Yes
Are you starting the service with sudo systemctl start voice-recognizer or by enabling it and restarting the Pi? - Running sudo systemctl start voice-recognizer
Are you passing the -I/--input-device or -O/--output-device parameters which wouldn't be used by the service? - No, simply running src/main.py works in dev console and I checked the systemd file and it uses no flags as you mentioned.
When it starts as a service, what errors do you see with sudo journalctl -n 20 -u voice-recognizer? I will get you that output.
from aiyprojects-raspbian.
dylandignan
commented on April 28, 2024
This gist (https://gist.github.com/diggy106/f71e55dd7db92de18dbef5baa838ccb6) has some output from the command you gave me and /var/log/syslog. I added a line in there to also log the aplay or arecord command.
Of note is in syslog while running as daemon I see the logs below, when running manually they are not present, nor are then in stdout.
May 11 18:37:45 raspberrypi kernel: [ 889.253753] voicehat-codec voicehat-codec: Enabling audio amp...
May 11 18:37:50 raspberrypi kernel: [ 894.003338] voicehat-codec voicehat-codec: Disabling audio amp...
I've also attached videos to show that check_audio.py and src/main.py works with no flags, but the daemon fails to play audio on the bluetooth speaker. I can see that it is picking up my microphone.
check_audio.py - https://youtu.be/2D2UsFHHMkE
src/main.py - https://youtu.be/Ab-Yi2_1cUQ
daemon - https://youtu.be/5fWBv0U7-h8
from aiyprojects-raspbian.
drigz
commented on April 28, 2024
@diggy106 Thanks for the videos, I have a much better idea what's going on now (but still no idea what's causing it!)
You could try running aplay -v /usr/share/sounds/alsa/Front_Center.wav on the console to show what device ALSA is using there, then patching audio.py to change '-q' to '-v' to see that output when the voice-recognizer is running as a daemon.
I'm surprised that enabling/disabling audio amp is in the syslog, since you're not using the Voice HAT. Have you edited your /boot/config.txt? If you have any of the lines dtoverlay=i2s-mmap, dtoverlay=googlevoicehat-soundcard or dtparam=i2s=on in there, you could try removing them and seeing if it fixes your problem.
from aiyprojects-raspbian.
dylandignan
commented on April 28, 2024
I did not change /boot/config.txt myself. I do see the lines you mentioned in that file. Seems to come from https://github.com/google/aiyprojects-raspbian/blob/master/scripts/configure-driver.sh which I assume was run as part of the image.
pi@raspberrypi:~ $ sudo grep dtoverlay=i2s-mmap /boot/config.txt
dtoverlay=i2s-mmap
pi@raspberrypi:~ $ sudo grep dtoverlay=googlevoicehat-soundcard /boot/config.txt
dtoverlay=googlevoicehat-soundcard
pi@raspberrypi:~ $ sudo grep dtparam=i2s=on /boot/config.txt
dtparam=i2s=on
Below is the output from aplay -v.
pi@raspberrypi:~ $ aplay -v /usr/share/sounds/alsa/Front_Center.wav
Playing WAVE '/usr/share/sounds/alsa/Front_Center.wav' : Signed 16 bit Little Endian, Rate 48000 Hz, Mono
ALSA <-> PulseAudio PCM I/O Plugin
Its setup is:
stream : PLAYBACK
access : RW_INTERLEAVED
format : S16_LE
subformat : STD
channels : 1
rate : 48000
exact rate : 48000 (48000/1)
msbits : 16
buffer_size : 24000
period_size : 6000
period_time : 125000
tstamp_mode : NONE
period_step : 1
avail_min : 6000
period_event : 0
start_threshold : 24000
stop_threshold : 24000
silence_threshold: 0
silence_size : 0
boundary : 1572864000
Here is syslog from the daemon with aplay -v
May 13 14:02:06 raspberrypi python3[1658]: [2017-05-13 14:02:06,059] INFO:main:listening...
May 13 14:02:06 raspberrypi python3[1658]: [2017-05-13 14:02:06,062] INFO:main:recognizing...
May 13 14:02:09 raspberrypi python3[1658]: [2017-05-13 14:02:09,364] INFO:speech:event_type: 1
May 13 14:02:09 raspberrypi python3[1658]: [2017-05-13 14:02:09,370] INFO:main:thinking...
May 13 14:02:09 raspberrypi python3[1658]: [2017-05-13 14:02:09,373] INFO:speech:transcript: test
May 13 14:02:09 raspberrypi python3[1658]: [2017-05-13 14:02:09,590] INFO:main:Playing 5.1630 seconds of audio...
May 13 14:02:09 raspberrypi python3[1658]: [2017-05-13 14:02:09,591] INFO:audio:aplay: self._cmd: ['aplay', '-v', '-t', 'raw', '-D', 'default', '-c', '1', '-f', 's16', '-r', '16000']
May 13 14:02:09 raspberrypi python3[1658]: Playing raw data 'stdin' : Signed 16 bit Little Endian, Rate 16000 Hz, Mono
May 13 14:02:09 raspberrypi python3[1658]: Plug PCM: Rate conversion PCM (48000, sformat=S16_LE)
May 13 14:02:09 raspberrypi python3[1658]: Converter: libspeex (builtin)
May 13 14:02:09 raspberrypi python3[1658]: Protocol version: 10002
May 13 14:02:09 raspberrypi python3[1658]: Its setup is:
May 13 14:02:09 raspberrypi python3[1658]: stream : PLAYBACK
May 13 14:02:09 raspberrypi python3[1658]: access : RW_INTERLEAVED
May 13 14:02:09 raspberrypi python3[1658]: format : S16_LE
May 13 14:02:09 raspberrypi python3[1658]: subformat : STD
May 13 14:02:09 raspberrypi python3[1658]: channels : 1
May 13 14:02:09 raspberrypi python3[1658]: rate : 16000
May 13 14:02:09 raspberrypi python3[1658]: exact rate : 16000 (16000/1)
May 13 14:02:09 raspberrypi python3[1658]: msbits : 16
May 13 14:02:09 raspberrypi python3[1658]: buffer_size : 8000
May 13 14:02:09 raspberrypi python3[1658]: period_size : 2000
May 13 14:02:09 raspberrypi python3[1658]: period_time : 125000
May 13 14:02:09 raspberrypi kernel: [ 416.025835] voicehat-codec voicehat-codec: Enabling audio amp...
May 13 14:02:09 raspberrypi python3[1658]: tstamp_mode : NONE
May 13 14:02:09 raspberrypi python3[1658]: period_step : 1
May 13 14:02:09 raspberrypi python3[1658]: avail_min : 2000
May 13 14:02:09 raspberrypi python3[1658]: period_event : 0
May 13 14:02:09 raspberrypi python3[1658]: start_threshold : 8000
May 13 14:02:09 raspberrypi python3[1658]: stop_threshold : 8000
May 13 14:02:09 raspberrypi python3[1658]: silence_threshold: 0
May 13 14:02:09 raspberrypi python3[1658]: silence_size : 0
May 13 14:02:09 raspberrypi python3[1658]: boundary : 524288000
May 13 14:02:09 raspberrypi python3[1658]: Slave: Route conversion PCM (sformat=S32_LE)
May 13 14:02:09 raspberrypi python3[1658]: Transformation table:
May 13 14:02:09 raspberrypi python3[1658]: 0 <- 0
May 13 14:02:09 raspberrypi python3[1658]: 1 <- 0
May 13 14:02:09 raspberrypi python3[1658]: Its setup is:
May 13 14:02:09 raspberrypi python3[1658]: stream : PLAYBACK
May 13 14:02:09 raspberrypi python3[1658]: access : MMAP_INTERLEAVED
May 13 14:02:09 raspberrypi python3[1658]: format : S16_LE
May 13 14:02:09 raspberrypi python3[1658]: subformat : STD
May 13 14:02:09 raspberrypi python3[1658]: channels : 1
May 13 14:02:09 raspberrypi python3[1658]: rate : 48000
May 13 14:02:09 raspberrypi python3[1658]: exact rate : 48000 (48000/1)
May 13 14:02:09 raspberrypi python3[1658]: msbits : 16
May 13 14:02:09 raspberrypi python3[1658]: buffer_size : 24000
May 13 14:02:09 raspberrypi python3[1658]: period_size : 6000
May 13 14:02:09 raspberrypi python3[1658]: period_time : 125000
May 13 14:02:09 raspberrypi python3[1658]: tstamp_mode : NONE
May 13 14:02:09 raspberrypi python3[1658]: period_step : 1
May 13 14:02:09 raspberrypi python3[1658]: avail_min : 6000
May 13 14:02:09 raspberrypi python3[1658]: period_event : 0
May 13 14:02:09 raspberrypi python3[1658]: start_threshold : 24000
May 13 14:02:09 raspberrypi python3[1658]: stop_threshold : 24000
May 13 14:02:09 raspberrypi python3[1658]: silence_threshold: 0
May 13 14:02:09 raspberrypi python3[1658]: silence_size : 0
May 13 14:02:09 raspberrypi python3[1658]: boundary : 1572864000
May 13 14:02:09 raspberrypi python3[1658]: Slave: Hardware PCM card 0 'snd_rpi_googlevoicehat_soundcar' device 0 subdevice 0
May 13 14:02:09 raspberrypi python3[1658]: Its setup is:
May 13 14:02:09 raspberrypi python3[1658]: stream : PLAYBACK
May 13 14:02:09 raspberrypi python3[1658]: access : MMAP_INTERLEAVED
May 13 14:02:09 raspberrypi python3[1658]: format : S32_LE
May 13 14:02:09 raspberrypi python3[1658]: subformat : STD
May 13 14:02:09 raspberrypi python3[1658]: channels : 2
May 13 14:02:09 raspberrypi python3[1658]: rate : 48000
May 13 14:02:09 raspberrypi python3[1658]: exact rate : 48000 (48000/1)
May 13 14:02:09 raspberrypi python3[1658]: msbits : 32
May 13 14:02:09 raspberrypi python3[1658]: buffer_size : 24000
May 13 14:02:09 raspberrypi python3[1658]: period_size : 6000
May 13 14:02:09 raspberrypi python3[1658]: period_time : 125000
May 13 14:02:09 raspberrypi python3[1658]: tstamp_mode : NONE
May 13 14:02:09 raspberrypi python3[1658]: period_step : 1
May 13 14:02:09 raspberrypi python3[1658]: avail_min : 6000
May 13 14:02:09 raspberrypi python3[1658]: period_event : 0
May 13 14:02:09 raspberrypi python3[1658]: start_threshold : 24000
May 13 14:02:09 raspberrypi python3[1658]: stop_threshold : 24000
May 13 14:02:09 raspberrypi python3[1658]: silence_threshold: 0
May 13 14:02:09 raspberrypi python3[1658]: silence_size : 0
May 13 14:02:09 raspberrypi python3[1658]: boundary : 1572864000
May 13 14:02:09 raspberrypi python3[1658]: appl_ptr : 0
May 13 14:02:09 raspberrypi python3[1658]: hw_ptr : 0
May 13 14:02:14 raspberrypi kernel: [ 421.275435] voicehat-codec voicehat-codec: Disabling audio amp...
May 13 14:02:14 raspberrypi python3[1658]: [2017-05-13 14:02:14,911] INFO:main:ready...
I'll try removing the lines you've mentioned and let you know.
from aiyprojects-raspbian.
dylandignan
commented on April 28, 2024
I was finally able to get it working with the daemon by following some of the instructions in that Android Authority article linked above. I am going to stick with the 3.5mm setup as it is more reliable than the bluetooth connection and don't have to worry about auto-connecting on startup.
/etc/asound.conf
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"
}
}
Added this to the bottom of my /boot/config.txt
# Enable audio (loads snd_bcm2835)
dtparam=audio=on
#dtoverlay=i2s-mmap
#dtoverlay=googlevoicehat-soundcard
syslog
May 13 14:32:47 raspberrypi python3[1244]: Plug PCM: Hardware PCM card 0 'bcm2835 ALSA' device 0 subdevice 0
May 13 14:32:47 raspberrypi python3[1244]: Its setup is:
May 13 14:32:47 raspberrypi python3[1244]: stream : PLAYBACK
May 13 14:32:47 raspberrypi python3[1244]: access : RW_INTERLEAVED
May 13 14:32:47 raspberrypi python3[1244]: format : S16_LE
May 13 14:32:47 raspberrypi python3[1244]: subformat : STD
May 13 14:32:47 raspberrypi python3[1244]: channels : 1
May 13 14:32:47 raspberrypi python3[1244]: rate : 16000
May 13 14:32:47 raspberrypi python3[1244]: exact rate : 16000 (16000/1)
May 13 14:32:47 raspberrypi python3[1244]: msbits : 16
May 13 14:32:47 raspberrypi python3[1244]: buffer_size : 8000
May 13 14:32:47 raspberrypi python3[1244]: period_size : 2000
May 13 14:32:47 raspberrypi python3[1244]: period_time : 125000
May 13 14:32:47 raspberrypi python3[1244]: tstamp_mode : NONE
May 13 14:32:47 raspberrypi python3[1244]: period_step : 1
May 13 14:32:47 raspberrypi python3[1244]: avail_min : 2000
May 13 14:32:47 raspberrypi python3[1244]: period_event : 0
May 13 14:32:47 raspberrypi python3[1244]: start_threshold : 8000
May 13 14:32:47 raspberrypi python3[1244]: stop_threshold : 8000
May 13 14:32:47 raspberrypi python3[1244]: silence_threshold: 0
May 13 14:32:47 raspberrypi python3[1244]: silence_size : 0
May 13 14:32:47 raspberrypi python3[1244]: boundary : 2097152000
May 13 14:32:47 raspberrypi python3[1244]: appl_ptr : 0
May 13 14:32:47 raspberrypi python3[1244]: hw_ptr : 0
May 13 14:32:49 raspberrypi python3[1244]: [2017-05-13 14:32:49,227] INFO:main:ready...
from aiyprojects-raspbian.
devilmind
commented on April 28, 2024
Hi,
working fine what you have suggested here but soon after reboot on my /home/pi/.asoundrc are appendend those line:
pcm.!default {
type hw
card 1
}
ctl.!default {
type hw
card 1
}
what is appending those line on reboot?
I need to delet them to get it worked again.
Can you help me please?
Thanks a lot.
from aiyprojects-raspbian.
Related Issues (20)
- AIY python package version HOT 2
- GET request in Robot framework using AWS credentials HOT 4
- say and play_wav failing after invoking CloudSpeechClient client.recognize method
- AYI-vision: Unit joy_dection_demo.service could not be found HOT 3
- AIY-voice: problems encountered by assistant_grpc_demo.py after some time of use HOT 3
- Python package install fails on raspbian
- aiy-python-wheels fails to get installed on raspbian bullseye on rpi4 HOT 4
- Pico TTS not installable (Bullseye)
- Voice Bonnet drivers not installed (HACKING.md instructions) HOT 1
- No sound input/output on Raspbian Bullseye HOT 1
- Huge delay in voice recording. HOT 1
- 500 error code HOT 3
- No microphone as user Pi on version 20210402
- Not working with Raspbian Bullseye HOT 1
- Voice Bonnet not working with Pi 4 HOT 1
- Pi Zero 2 W?
- Voice Kit Google - Failed (src/examples/voice/assistant_grpc_demo.py)
- post-installation script subprocess returned error exit status 6 HOT 3
- Voice kit google (/lib/arm-linux-gnueabihf/libc.so.6: version 'GLIBC_2.34' not found (required by /lib/arm-linux-gnueabihf/libgcc_s.so.1))
- Sound driver for AIY Voice Kit V1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from aiyprojects-raspbian.