OpenTTS Setup

Back to the TTS Guide

OpenTTS is an open-source TTS server that runs on your own computer using Docker. No account, no API key, no data leaving your machine. Setup takes about 2 minutes if you already have Docker installed.

Note: OpenTTS uses older neural and rule-based engines. Voice quality is functional but noticeably more robotic than ElevenLabs, Google Cloud, or KittenTTS. It may be removed in a future release if better local options emerge.

Step 1: Install Docker

If you don’t have Docker yet:

• Linux (Ubuntu/Debian): sudo apt install docker.io
• Linux (Fedora): sudo dnf install docker
• macOS / Windows: Download Docker Desktop

Step 2: Start the OpenTTS Server

Open a terminal and run:

docker run -d -p 5500:5500 --name opentts synesthesiam/opentts:en

This downloads the English voice pack (~1.5 GB on first run) and starts the server in the background. It runs until you stop it.

Other Languages

Other language packs are available: de, fr, es, ru, nl, sv, it, and more. For all languages (larger download):

docker run -d -p 5500:5500 --name opentts synesthesiam/opentts:all

CJK Warning: OpenTTS works best with European languages. Japanese, Chinese, and Korean text will not be pronounced correctly. For CJK languages, use ElevenLabs or Google Cloud instead.

Step 3: Configure En Parlant~

1. Open En Parlant~ and go to Settings (gear icon) > Sound tab
2. Scroll down to the TTS section
3. Set TTS Provider to OpenTTS (Self-Hosted)
4. Confirm the OpenTTS Server URL is http://localhost:5500
5. The TTS Voice dropdown will populate with available voices. Try a larynx voice (like harvard) for the best quality
6. Set Text-to-Speech to On
7. Click the Test button next to the voice selector

You should hear a chess move spoken aloud.

Voice Quality Guide

Voices come from several engines bundled in OpenTTS. From best to most basic:

1. Larynx — neural, most natural sounding
2. Coqui-TTS — neural, multi-speaker
3. MaryTTS — Java-based, decent quality
4. Festival — traditional synthesis
5. eSpeak — robotic but fast

The voice dropdown shows the engine name in parentheses so you can pick accordingly.

Managing the Server

The OpenTTS container runs in the background. Common commands:

# Stop the server
docker stop opentts

# Start it again
docker start opentts

# Remove it entirely
docker rm -f opentts

# Check if it's running
docker ps | grep opentts

Troubleshooting

• “Connection refused” error? The Docker container may not be running. Run docker ps to check. If it’s not listed, start it with docker start opentts or re-run the docker run command from Step 2.
• No voices in dropdown? Make sure the container is running and accessible at http://localhost:5500. You can test by opening that URL in your browser — you should see the OpenTTS web interface.
• Voices sound very robotic? Try switching to a Larynx voice — they’re the highest quality option in OpenTTS. If you want better quality overall, consider upgrading to KittenTTS, Google Cloud, or ElevenLabs.