Error setting up local voice pipeline - piper, whisper, openwakeword

Configuration Voice Assistant
1

So I installed piper, whisper and openwakeword

  piper:
    image: rhasspy/wyoming-piper
    container_name: piper
    hostname: piper
    restart: unless-stopped
    user: "0"
    environment:
      TZ: Europe/Nicosia
    ports:
      - 10200:10200
    volumes:
      - ./config/piper:/data
    command: --voice en-gb-alan-low

  whisper:
    image: rhasspy/wyoming-whisper
    container_name: whisper
    hostname: whisper
    restart: unless-stopped
    user: "0"
    environment:
      TZ: Europe/Nicosia
    ports:
      - 10300:10300
    volumes:
      - ./config/whisper:/data
    command: --data-dir /data --model tiny-int8 --language en

  openwakeword:
    image: rhasspy/wyoming-openwakeword
    container_name: openwakeword
    hostname: openwakeword
    restart: unless-stopped
    user: "0"
    environment:
      TZ: Europe/Nicosia
    ports:
      - 10400:10400
    volumes:
      - ./config/openwakeword:/custom
    command: --preload-model 'ok_nabu' --custom-model-dir /custom

Added them under the Wyoming integration

image
image932×395 13 KB

And headed over to voice assistants to add them.

image
image455×884 43.5 KB

But when I click create, I get this error

image
image446×453 21.5 KB

Any ideas why?

2

It seems this might be a bug with Home Assistant. It occurs even with official integrations like Elevenlabs.

3

Tap the language selection field for stt and tts again and select an available option.
The voice selection field should also be displayed if everything is set up correctly.