Rhasspy offline voice assistant toolkit

Share your Projects! Custom Integrations
606

Running as user pi doesnt change this, still the same error with our without the “–config” flag.
I followed tips in the espeak thread and could get rid of the warnings.
i installed pip3 for this, python and python3 (3.5) were installed before.

without the conifg in place i get the following:

root@raspberrypi:/home/pi# hermes-audio-player
hermes-audio-player 0.1.1
Configuration file /etc/hermes-audio-server.json not found.

if the json has a formatting error i get a JSONDecodeError. Only if the file is correctly formatted and in place im getting the No such file or directory error … very strange

607

Ok, I think I’m beginning to see what is going on. I suspect the problem is not with the configuration file, because the error comes after the message Connected to audio output default., and at that moment the configuration file is already read. But it’s before the Connected to MQTT broker message appears. Maybe double-check your MQTT connection settings in the configuration file? What is your (anonymized) /etc/hermes-audio-server.json configuration anyway?

I’ll have to improve the error handling and logging, that’s on my TODO list…

608

thanks for taking the time to help me, but i still cant get it working.
I double checked the mqtt settings and everything is correct:

{
    "site": "default",
    "mqtt": {
        "host": "192.168.1.8",
        "port": 1883,
        "authentication": {
            "username": "mqtt",
            "password": "mypassword@mypassword"
        },
        "tls": {
            "ca_certificates": "",
            "client_certificate": "",
            "client_key": ""
        }
    },
    "vad": {
        "mode": 0,
        "silence": 2,
        "status_messages": false
    }
}

I have a @ symbol in my password, could this be the problem?

609

It could be. Can you try connecting to another MQTT broker, e.g. by installing mosquitto on the machine and using it without authentication or with a simple password? Or temporarily change the password on your MQTT broker?

610

Ok im using mosquitto on my home server and created another simple username and password (rhasspy:rhasspy). I tried this user with another device and it authenticates fine.
Sadly i still get the same error with hermes.

611

Ok, thanks for trying this. I still don’t understand what’s going on, but I’ll do some extra testing and I’ll add better error messages and logging for the next version. For now I don’t have a solution for your problem, but these additions should (help) solve your problem, so stay tuned :slight_smile:

612

No problem, glad i can help testing this.

613

I’ve tried finding an answer on this question for weeks now, but I can’t seem to find a satisfactory solution on this.

I’m running Hass.io on an Intel NUC, hidden away in a server closet. I would like to use voice recognition in my house, which means that I’ll need microphones everywhere. I’m curious if this is possible. So far it seems I can only choose one wired microphone as input.

Is it possible to use multiple microphones running on a small wireless device like a ESP32 to send commands to Rhasspy?

It would be great to be able to just talk anywhere and see the house handle it.

614

Did you find the remote audio projects in Rhasspy’s hardware documentation? However, I don’t know if Rhasspy supports multiple audio devices yet.

615

I’ve seen it, but I’ve not yet tried it. Currently I’m trying to find answers on all my questions, so that I have the correct hardware.

I’m also not sure what microphone to use. I’ve seen some of these “Mic Arrays” but I find them quite expensive and it should be possible to use some smaller out of the box microphones I suppose.

616

Yes, that is possible. Your device needs to be able to send the data via MQTT to a broker.
What koan says, remote audio documentation is a good source.

For a mic, you also could use a Pi Zero with a mic and use koans audio server.

617

Updating addon on to 2.17 gives this error:

usage: Rhasspy [-h] --profile PROFILE [–host HOST] [–port PORT]
[–system-profiles SYSTEM_PROFILES]
[–user-profiles USER_PROFILES] [–set SET SET] [–ssl SSL SSL]
Rhasspy: error: unrecognized arguments:

618

Great, I’m gonna look into it. Thank you!

619

Did the bump to 2.17.1 help? I don’t have a hass.io system readily available to test on.

620

Yes! But it asked me to download profile again and keeps asking :smiley:

621

Crumbs :frowning:

622

Only for non EN though, I have set the language to NL

623

OK, this should be fixed now. I re-wrote the download piece of Rhasspy for the latest version, and forgot to create intermediary directories.

624

Version 2.17 is now out (with some bug fixes)! I’m trying to focus the new few releases of Rhasspy on “quality of life” improvements instead of new features – fixing bugs and improving the user experience.

Here’s what new:

  • Added the porcupine wake word system (way better than the others)
    • There are a bunch of pre-trained keyword files available, but you have to pick the right one for your platform (desktop/laptop vs. raspberry pi)
    • You can train your own keyword, but it only lasts 30 days (you can just train it again, though)
  • Added a “Problems” tab to the web interface that tries to help with common Rhasspy problems. It only shows up if a problem is detected at start up. I’ll add more checks over time.
  • Re-wrote the profile download logic to be triggered by specific profile settings. Files are now only downloaded when absolutely needed.
  • Lots of bug fixes

One thing I forgot to mention from the last release: language model mixing. You can “open up” your voice commands to include general words/phrases from your language without adding it to sentences.ini. This makes training/recognition take a lot longer, but gives you a lot more flexibility in how you say things.

1 Like
625

Great new features and indeed a good idea to stablize now :slight_smile:
But I am already on 2.17?

Sorry, I see 2.17.2 now :slight_smile: