# shivammehta25/Matcha-TTS

[ICASSP 2024] 🍵 Matcha-TTS: A fast TTS architecture with conditional flow matching

Repository: https://github.com/shivammehta25/Matcha-TTS
Canonical: https://ross.abutalabs.com/products/matcha-tts
Homepage: https://shivammehta25.github.io/Matcha-TTS/
Language: Jupyter Notebook
License: MIT
License Family: permissive
Topics: deep-learning, flow-matching, machine-learning, non-autoregressive, probabilistic, probabilistic-machine-learning, text-to-speech, tts, tts-api, tts-engines, diffusion-model, diffusion-models
Last push: 2026-08-17T22:39:44+00:00

## Health v2 (maintenance only)
Score: 62/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 8, longevity 78
- inputs: {"age_days": 1093, "days_push": 16, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1349, forks 214 (observed 2026-08-28T04:04:27.812623+00:00)

## What it is
Matcha-TTS is a PyTorch-based text-to-speech system that uses conditional flow matching for fast, non-autoregressive speech synthesis. It ships as a pip-installable package with a CLI, Gradio app, and pre-trained models.

## Use cases
- convert text to natural-sounding speech
- fast neural text-to-speech synthesis
- generate speech audio from text on the command line
- run a lightweight TTS model with a small memory footprint
- experiment with flow matching for speech synthesis
- self-host a TTS API or web demo

## When to choose
- you need fast, probabilistic TTS with natural-sounding output
- you want a compact model that runs quickly without autoregressive decoding
- you want a pip-installable TTS with CLI and Gradio interfaces and pre-trained checkpoints

## When to avoid
- you need voice cloning or fine-grained speaker control out of the box
- you need production multi-speaker TTS with extensive language coverage
- you prefer mature, widely-deployed TTS engines with large ecosystems

## Facets
- artifact type: library
- maturity: active
- function: tts, machine-learning, deep-learning, cli
- domain: speech-processing, machine-learning, deep-learning
- platform: python, cli, cross-platform
- tags: text-to-speech, flow-matching, diffusion-models, non-autoregressive, speech-synthesis, pytorch, pretrained-models, natural-language-processing, gpu

## Member repositories
- shivammehta25/Matcha-TTS (main) score 62

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:27.812623+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:42:23.194501+00:00, confidence not recorded.
  - readme: https://github.com/shivammehta25/Matcha-TTS (fetched 2026-08-28T04:04:27.812623+00:00, sha 9e17b65ef497)
  - homepage: https://shivammehta25.github.io/Matcha-TTS/ (fetched 2026-08-29T12:01:22.605046+00:00, sha cbc240b1e957)
- Data as of 2026-08-30T08:39:29.467469+00:00.