Getting Started with Coqui TTS: Docker Setup and Voice Synthesis Guide

Introduction

Coqui TTS is the leading open-source text-to-speech toolkit, and getting it running on your own infrastructure is surprisingly straightforward thanks to its official Docker image. This guide walks you through deploying Coqui TTS, generating your first audio, and exploring voice cloning.

Prerequisites

Docker installed (Docker Desktop or Docker Engine)
At least 2GB of free RAM (8GB recommended for larger models)
A terminal or command prompt

Step 1: Pull and Run the Docker Container

Start by pulling the official CPU image and running the TTS server:

docker run --rm -it -p 5002:5002 ghcr.io/coqui-ai/tts-cpu

The server starts on port 5002. Access the web demo at http://localhost:5002.

Step 2: Generate Speech via API

Once the server is running, generate speech with curl:

curl -X POST "http://localhost:5002/api/tts" \
  -H "Content-Type: application/json" \
  -d '{"text": "Hello, this is Coqui TTS speaking!","speaker_id": "default"}' \
  --output speech.wav

Step 3: Try Voice Cloning

Coqui TTS supports voice cloning with just a few seconds of reference audio:

curl -X POST "http://localhost:5002/api/tts" \
  -H "Content-Type: application/json" \
  -d '{"text":"This sounds like the reference.","speaker_id":"cloned","reference_wav":"/path/to/reference.wav"}' \
  --output cloned.wav

Running with GPU

For lower latency, use the GPU image with NVIDIA GPUs:

docker run --rm -it --gpus all -p 5002:5002 ghcr.io/coqui-ai/tts

Docker Compose for Production

Create a docker-compose.yml:

version: "3.8"
services:
  coqui-tts:
    image: ghcr.io/coqui-ai/tts-cpu:latest
    restart: unless-stopped
    ports:
      - "5002:5002"
    volumes:
      - ./data/coqui-tts:/data

Then run docker compose up -d for a persistent deployment.

Conclusion

Coqui TTS makes it easy to add high-quality speech synthesis to any application. With Docker, you can be up and running in minutes.