MisoLabsAI/MisoTTS

Health v2 · maintenance only

Flags: no_releases young no_license

Adoption not part of the score

3224 stars · 328 forks observed · 2026-08-28

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

Miso TTS 8B is an open-source text-to-speech model based on an RVQ Transformer architecture with a Llama 3.2-style 8B backbone, designed for highly emotive conversational speech generation. The repository provides inference code, model definition, and setup instructions for running the model locally in Python.

Use cases

• generate expressive speech from text
• run a text-to-speech model locally
• synthesize conversational dialogue audio
• create emotive voiceovers from scripts
• convert text to wav audio with a large TTS model
• build a voice assistant with natural speech

When to choose

• you need high-quality, emotive English speech synthesis
• you want to run a state-of-the-art 8B TTS model on your own hardware
• you need dialogue-style conversational audio generation
• you want an open model with local inference instead of a cloud API

When to avoid

• you need multilingual or non-English speech (English only)
• you lack a capable GPU or the memory for an 8B model
• you need a lightweight, low-latency TTS for embedded or edge devices
• you need a permissively licensed model (license is custom/unspecified)

Facets

library · maturity active

tts machine-learning llm-inference audio-processing speech-processing artificial-intelligence deep-learning python text-to-speech voice-synthesis conversational-ai rvq-transformer llama-backbone speech-generation huggingface audio linux macos gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="MisoLabsAI/MisoTTS")