Free local text to speech for Open WebUI

1Start the speech server

With Docker, which includes FFmpeg:

docker run -d --name loudness-tts -p 8880:8880 -v loudness-tts:/data ghcr.io/loudness-ai/loudness-tts

Or with Python, listening on all network interfaces so that the Open WebUI container can reach it:

python -m pip install loudness-tts
loudness-tts serve --host 0.0.0.0 --port 8880 --api-key choose-a-secret

Check that it answers: http://127.0.0.1:8880/health in your browser shows "status": "ok".

2Open the audio settings

In Open WebUI, open Admin Panel → Settings → Audio.

3Fill in the text-to-speech settings

• Text-to-Speech Engine: OpenAI
• API Base URL: http://host.docker.internal:8880/v1 when Open WebUI runs in Docker, or http://127.0.0.1:8880/v1 when it runs directly on the same computer
• API Key: the secret you chose, or any text if the server has no key
• TTS Model: tts-1
• TTS Voice: nova, or a voice id

Then select Save. On Linux, start the Open WebUI container with --add-host=host.docker.internal:host-gateway so that host.docker.internal reaches your computer.

4Test it

Start a new chat, send a message and press the speaker icon under the answer. The first answer can take a few seconds longer while the voice downloads.

5Choose voices and languages

OpenAI voice names map to strong English (US) voices: nova, shimmer, coral, sage and marin are female; echo, onyx, ash, ballad, verse and cedar are male; alloy and fable are the top voices overall. For another language, enter a voice id as the TTS voice, such as de_DE-mls-medium-s162, or start the server with --language de-DE. loudness-tts voices --language fr lists the ids; listen first in the browser demo or the Piper Voice Explorer.

6Optional: run both with Docker Compose

To start Open WebUI and the speech server together, with the audio settings filled in on the first start:

services:
  open-webui:
    image: ghcr.io/open-webui/open-webui:main
    ports: ["3000:8080"]
    volumes: ["open-webui:/app/backend/data"]
    environment:
      AUDIO_TTS_ENGINE: openai
      AUDIO_TTS_OPENAI_API_BASE_URL: http://loudness-tts:8880/v1
      AUDIO_TTS_OPENAI_API_KEY: local
      AUDIO_TTS_MODEL: tts-1
      AUDIO_TTS_VOICE: nova
  loudness-tts:
    image: ghcr.io/loudness-ai/loudness-tts
    volumes: ["loudness-tts:/data"]
volumes:
  open-webui:
  loudness-tts:

Open WebUI applies these values on its first start; after that, change them in the Admin Panel. Details are in Open WebUI’s documentation.