XTTS or F5-TTS: which voice cloning model to pick
Both models clone a voice from a short sample and both run on a regular computer without a GPU. You may only clone a voice with the written consent of the person who owns it. XTTS covers 17 languages including Russian, but Coqui has shut down and development stopped at v2 in 2023-10. F5-TTS is newer (last update 2025-03), officially English and Chinese, but the community has released fine-tunes for other languages, Russian among them. Commercial use is closed for both: XTTS under the Coqui Public Model License, the official F5-TTS weights under CC BY-NC 4.0, so a product needs a different option.
Comparison based on catalog data
| Parameter | Coqui XTTS | F5-TTS |
|---|---|---|
| Category | Text to speech | Text to speech |
| Developer | Coqui, Germany | Shanghai Jiao Tong University and partners, China |
| Releases | Sep 2023 – Oct 2023 | Oct 2024 – Mar 2025 |
| Sizes | about 470M | about 340M |
| Hardware | Laptop | Laptop |
| Commercial use | Non-commercial only | Non-commercial only |
| License | Coqui Public Model License, non-commercial use only | Code MIT, official weights CC BY-NC 4.0, non-commercial |
| Russian | Supported | Not supported |
| Ollama | No | No |
| Without GPU | Yes | No |
| Tasks |
|
|
Choose Coqui XTTS if
- You need Russian out of the box, without hunting for community fine-tunes
- The project is research or education, where a non-commercial license is fine
- Language coverage matters: XTTS lists 17 languages
Choose F5-TTS if
- You want a living project: F5-TTS was updated in 2025, XTTS stopped in 2023
- You are ready to use community fine-tunes for your language
- You want a smaller model: about 340M against 470M for XTTS
Other comparisons
- F5-TTS or Fish Speech: which voiceover model to pick
- Silero or Piper: which to choose for business
- Piper or Kokoro: which text-to-speech model to pick
- GigaAM or Vosk: which speech recognition to pick
- Qwen or GigaChat: which to choose for business
- Qwen or Llama: which to choose for business
- GigaChat or YandexGPT: which to choose for business
- DeepSeek or Qwen: which to choose for business
- Gemma or Llama: which to choose for business
- Mistral or Llama: which to choose for business
- Whisper or GigaAM: which to choose for business
- Whisper or Parakeet: which to choose for business
Need a model for your task?
An open model can run on your own server: data stays in-house, there is no per-request fee, and the model can be fine-tuned on your documents.
- SelectThe model and size for your task and hardware budget
- DeployOn your server or in a closed network, with an API
- Fine-tuneOn your data, or connect a knowledge base
- IntegrateInto your CRM, ERP, bot, website or team chat