Coqui XTTS-v2
Once the default open cloning TTS (17 languages, <200 ms streaming). Coqui the company is gone; the code lives on in the Idiap fork (coqui-tts on PyPI), but the weights remain under the non-commercial Coqui Public Model License.
Overview
Best for: Legacy/non-commercial projects already built on it.
At a glance
Project claims under 200 ms streaming on GPU. Weights under the non-commercial Coqui Public Model License with no way to buy a commercial licence. Unmaintained since 2023.
Text plus 6 s+ reference audio
24 kHz
17
Zero-shot cloning
Project claim: can stream with <200 ms latency (GPU).
Wherever you deploy it
Your own deployment; no vendor data processing
GPU for real-time
Coqui Public Model License (non-commercial) for weights; code MPL-2.0 in the fork (not re-verified)
Features
- streaming inference
- cloning
- fine-tuning recipes
Pricing
| What | Price | Unit |
|---|---|---|
| Weights | $0 |
Self-hosted: cost is your GPU/CPU time, not per character
Free tier: Open weights
Source: huggingface.co
Setup
- pip install coqui-tts (Idiap fork).
- Load TTS('tts_models/multilingual/multi-dataset/xtts_v2') on GPU.
Endpoint
LocalAuthentication
None
Quick start python
# pip install coqui-tts (maintained Idiap fork)
from TTS.api import TTS
tts = TTS("tts_models/multilingual/multi-dataset/xtts_v2").to("cuda")
tts.tts_to_file(text="Hello from XTTS.", speaker_wav="reference.wav",
language="en", file_path="out.wav")Written from the current docs. Check the vendor's SDK version before you ship.
Warnings
No way to license commercially
Weights are under the non-commercial CPML and Coqui no longer exists to sell a commercial licence.
Weights unmaintained since 2023
Quality and robustness lag 2026 models such as Chatterbox, Qwen3-TTS or VoxCPM2.
Use the fork
The original coqui-ai/TTS repo is unmaintained; install coqui-tts from the Idiap fork.
Plus 3 warnings that apply to all open models APIs. See category warnings.
Limits
- CPML non-commercial
Models and products
| Name | Status |
|---|---|
| tts_models/multilingual/multi-dataset/xtts_v2 | Unmaintained weights (2023) |
Docs and sources
Docs
Sources used
Fork code licence.