MeloTTS Spanish
myshell-ai/MeloTTS-Spanish
published Feb 2024 · updated Mar 2024
MeloTTS Spanish is a text-to-speech model that generates high-quality Spanish speech from text.
specs
| Task | Text-to-Speech (TTS) |
| Architecture | VITS-based (VITS, VITS2, Bert-VITS2) |
| License | MIT |
about this model
myshell-ai/MeloTTS-Spanish is a text-to-speech model that produces natural, high-quality Spanish speech. It is part of the MeloTTS family developed by MyShell.ai in collaboration with MIT and Tsinghua University. The model supports CPU real-time inference, making it fast and accessible for production deployments without requiring a GPU.
Supported Languages
While the Spanish model is optimized for Spanish, the MeloTTS family covers multiple languages and accents, all accessible via the same architecture:
| Language | Example Audio |
|---|---|
| English (American) | Listen |
| English (British) | Listen |
| English (Indian) | Listen |
| English (Australian) | Listen |
| Spanish | Listen |
| French | Listen |
| Chinese (mixed EN) | Listen |
| Japanese | Listen |
| Korean | Listen |
The Chinese model additionally supports mixed Chinese-English output. All models are fast enough for CPU real-time inference.
Community Adoption
The MeloTTS-Spanish model has been downloaded over 71,000 times in the past month and is used by 17 Hugging Face Spaces, reflecting active real-world deployment.
License and Attribution
Licensed under MIT, free for commercial and non-commercial use. The official citation is @software{zhao2024melo, author = {Zhao, Wenliang and Yu, Xumin and Qin, Zengyi}, title = {MeloTTS: High-quality Multi-lingual Multi-accent Text-to-Speech}, url = {https://github.com/myshell-ai/MeloTTS}, year = {2023}}.
best for
- ·Spanish voiceovers for videos and presentations
- ·Accessible reading tools for Spanish text
- ·Real-time speech synthesis on CPU
FAQ
It accepts Spanish text and produces a WAV audio file of synthesized speech.
Yes, it is fast enough for CPU real-time inference.
It is released under the MIT License, allowing both commercial and non-commercial use.
Use the OpenAI-compatible endpoint with your API key, sending the text and receiving audio in response.
No, this specific model is trained only for Spanish. Other language variants are available separately.
We're benchmarking and onboarding MeloTTS Spanish as a hosted, OpenAI-compatible API. Sign in for free credit and be ready when it lands, or tell us you want it and we'll prioritize it.