myshell-ai/MeloTTS

Health v2 · maintenance only

Adoption not part of the score

7608 stars · 1066 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

MeloTTS is a high-quality multi-lingual text-to-speech library supporting English (multiple accents), Spanish, French, Chinese, Japanese, and Korean. It offers a Python API, CLI, and web UI, and is fast enough for CPU real-time inference.

Use cases

• convert text to natural-sounding speech in multiple languages
• generate Chinese speech mixed with English words
• run real-time TTS inference on CPU without a GPU
• build a voice assistant or chatbot with spoken replies
• narrate articles or documents as audio in different accents
• fine-tune TTS models on a custom voice dataset

When to choose

• you need multilingual TTS including East Asian languages
• you want fast CPU-only inference without heavy hardware
• you need a permissive MIT-licensed TTS engine you can self-host
• you want a Python API plus CLI and web UI out of the box

When to avoid

• you need voice cloning from short reference samples
• you require languages beyond the supported set
• you need the absolute highest-fidelity studio-grade synthesis
• you want a managed cloud TTS service with SLAs

Facets

library · maturity active

tts speech-recognition machine-learning cli speech-processing machine-learning python cross-platform cli text-to-speech multilingual cpu-inference voice-synthesis mit-license natural-language-processing audio

Member repositories

For agents

markdown · JSON · MCP: product_card(name="myshell-ai/MeloTTS")