MeloTTS myshell-ai/MeloTTS

Multilingual text-to-speech library from MyShell.ai and MIT covering English accents, Spanish, French, Chinese, Japanese and Korean.

Creator kit + Compare GitHub ↗ Install guide (official) ↗

MeloTTS at a glance

Checked 15 Oct 2026

MeloTTS is a VITS-based multilingual text-to-speech library from MyShell.ai with English (US, UK, Indian, Australian), Spanish, French, Chinese (mixed with English), Japanese and Korean voices. Code and Hugging Face weights are MIT, and it is described as fast enough for real-time CPU inference. Linux and macOS install natively; Windows is via Docker. No voice cloning or speech recognition. The repo has not been pushed to since December 2024.

Cite this pageRepoQuarry. “MeloTTS.” Checked 15 Oct 2026. https://repoquarry.dev/projects/melotts

How much of ElevenLabs does it replace?

1 of 4 jobs. Speech generation with fixed preset speakers per language; no voice cloning.

Speech generation Partial replacement

Voice cloning Not covered

Speech recognition Not covered

Voice agents Not covered

For voice cloning, speech recognition, voice agents, see other ElevenLabs alternatives. You may also need to build an interface, an API server and hosting yourself.

Facts, with sources

Every answer links to where we found it. We read the documentation; we did not install it.

Report a correction

QuestionAnswerSourceChecked

Runs on my machine? Yes, CPU or GPU (CUDA or MPS) github.com · install.md ↗ 1d ago

Code licence MIT This library is under MIT License, which means it is free for both commercial and non-commercial use. github.com · LICENSE ↗ 1d ago

Model weights licence MIT (MeloTTS weights on Hugging Face, all languages checked) huggingface.co · MeloTTS-English ↗ 1d ago

Commercial use Yes, MIT code and weights This library is under MIT License, which means it is free for both commercial and non-commercial use. github.com · MeloTTS#license ↗ 1d ago

Paid services needed? None required github.com · install.md ↗ 1d ago

Hardware CPU is enough for real-time inference; no RAM or VRAM numbers stated Fast enough for CPU real-time inference. github.com · MeloTTS#introduction ↗ 1d ago

Operating systems Linux and macOS native install (tested on Ubuntu 20.04); Windows via Docker To avoid compatibility issues, for Windows users and some macOS users, we suggest to run via Docker. github.com · install.md ↗ 1d ago

Setup effort Medium, install from source with pip -e plus a unidic download, or build the Docker image github.com · install.md ↗ 1d ago