F***ing excellent!!!
Agreedā¦great news. Hopefully I can revisit this next year after a few HA updates and get this going.
Wow, nice christmas gift !!!
So what happened to Mycroft?
Iāve enjoyed working with the people at Mycroft, but it was hard to say no to a job offer to work full time on my first love (Rhasspy)
Excellent news!
Iām trying to set up a PS3 Eye microphone in Rhasspy (Home Assistant addon) and have two problems:
-
Microphone only starts transmitting audio to rhasspy if I execute in HA console (in web ui):
ha audio default input --name "alsa_input..."
I found this way here. But for me it only works until reboot. -
The microphone does not hear well and to make it more sensitive, as I understand it, I need to configure asound.conf. But I donāt know where to place it. I tried creating it in all possible places:
- /etc/asound.conf
- ${HOME}/.asoundrc
- /usr/share/hassio/audio/asound
- asoundrc input string in addon config
pcm.jack {
type hw
card ALSA
}
pcm.array {
type hw
card CameraB409241
}
pcm.softvol {
type softvol
slave.pcm "jack"
control {
name Master
card 0
}
}
pcm.cap {
type plug
slave.pcm "array"
slave.channels 4
ttable {
0.0 15.0
1.1 15.0
}
}
pcm.!default {
type asym
playback.pcm "plug:softvol"
capture.pcm {
type plug
slave.pcm "cap"
}
}
But it either does nothing or it breaks the sound in Rhasspy. I need help.
I have a different kind of problemā¦ After I installed Rhasspy I only put 1 switch and 1 light in the slots and the buttons on my HA are now goneā¦
Anyone have any idea why that happened?
To the right of the Light should be a button for a Fan but itās no longer appearing since installing Rhasspyā¦ The fan was put in as a switchā¦ the light associated with it is the sameā¦button is now gone.
I found a way to run Rhasspy in a docker image with Home Assistant Supervised.
You can check the operation of the microphone and speaker with the following commands:
arecord -d 5 -D plughw:1,0 test.wav
aplay -D plughw:2,0 test.wav
By specifying instead of plughw:x,x devices of your host system, looking at their commands arecord -l
and aplay -l
.
The problem is that the hostās audio devices are busy with the Home Assistantās audio system. This issue is reported here: https://github.com/home-assistant/plugin-audio/issues/12
I used the solution from the penultimate post.
Once the microphone and speaker are available, /etc/asound.conf can be configured. I have it set up like this (I use PS3 eye):
pcm.array {
type hw
card CameraB409241 # microphone device name
}
pcm.array_gain {
type softvol
slave.pcm "array"
control {
name "PSEye"
count 2
card 0
}
min_dB -5.0
max_dB 30.0
}
pcm.cap {
type plug
slave.pcm "array_gain"
slave.channels 4
ttable {
0.0 15.0
1.1 15.0
}
}
pcm.speaker {
type plug
slave {
pcm {
type hw
card "KTUSBAUDIO" # speaker device name
device 0
}
}
}
pcm.!default {
type asym
capture.pcm "cap"
playback.pcm "speaker"
}
ctl.!default {
type hw
card 0
}
The path to asound.conf must be specified when running the docker image:
docker run -d -p 12101:12101 --name rhasspy --restart unless-stopped -v "$HOME/.config/rhasspy/profiles:/profiles" -v "/etc/localtime:/etc/localtime:ro" -v "/etc/asound.conf:/etc/asound.conf" --device /dev/snd:/dev/snd rhasspy/rhasspy:2.5.11 --user-profiles /profiles --profile en
With this configuration, everything works well. The only problem is, when the host is rebooted, the microphone becomes unavailable and its cable needs to be re-plugged. If anyone knows how this can be solved, I will be glad to advice.
Has anyone else noticed that since Home Assistant 2023.2.x some of the built-in intents donāt give spoken feedback through Rhasspy anymore? For me,
- HassTurnOn
- HassTurnOff
- HassToggle
- HassLightSet
are broken in that sense, whereas
- HassOpenCover
- HassCloseCover
are still working fine. The intent scripts Iāve written myself still work as intended too.
Rhasspy 3 developer preview is now available: Rhasspy 3 Developer Preview - Announcements - Rhasspy Voice Assistant
@synesthesiam just curious, is there any point in me continuing to play with Rhasspy, or should I just wait for HAās year of the voice to mature?
Playing with Rhasspy will get you āearly accessā to some things. For example, the Rhasspy 3 add-on already does local STT with Whisper, uses Assist, and does local TTS too. If you already have some Raspberry Piās for satellites and free, this might be the way to go for now.
I donāt know exactly what the HA side of things is going to end up like, but you can bet it will be more polished and streamlined than Rhasspy
Does anyone have a way to configure Rhasspy to send all audio playing to Sonos ?
When I set up Rhasspy I went the the automation route instead of the intent route and then I use automation like this:
id: Rhasspy Say the Date
alias: Rhasspy Say the Date
trigger:
platform: event
event_type: rhasspy_GetDate
action:
- service: script.sonos_say
data:
sonos_entity: media_player.living_room
volume: 0.5
message: " It is {{ states.sensor.date.state }} "
Has similar issue, solved it by omitting the content type, not sending binary data and pointing rhasspy to its local media folder path.
curl -X POST 'http://10.0.0.X:12101/api/play-wav' -d '/profiles/media/alerts/audio.wav'