MeloTTS-Korean is a text-to-speech model for the Korean language hosted on Hugging Face. It forms part of the MeloTTS multi-lingual text-to-speech library developed by MyShell.ai and is made available as open weights under the MIT license.
The model can be loaded and run through the Transformers library from Hugging Face. Users invoke it via a high-level pipeline for text-to-speech tasks or by directly instantiating the AutoModel class with automatic device mapping. Example code demonstrates initialization in a few lines, enabling local execution or integration into notebooks such as Google Colab and Kaggle.
MeloTTS supports multiple languages beyond Korean, with the broader library listing variants for several English accents as well as Spanish and French. The Korean-specific model is one implementation within this collection. The repository includes a model card with usage instructions, links to related notebooks, and community resources.
It is distributed for free under an MIT license with no pricing or paid tiers mentioned. The page provides direct access to model files, versions, and deployment options through Hugging Face infrastructure including inference providers.
MeloTTS Korean is a Voice, TTS & speech product. It focuses on generating natural-sounding Korean speech from text for applications and content creators. It is built as an open-source project for developers. MeloTTS Korean is open source under the MIT license. MeloTTS Korean is available on the web and API.
It is developed by MyShell.ai, and the product first shipped in 2024. Development happens publicly on GitHub with 7.5k stars. Key capabilities include text-to-speech, korean language support, and multi-speaker.
Latest indexed changes and source events
myshell-ai/MeloTTS-Korean verified by the PulseGate indexer
Other apps tracked under the same category.
