Matcha-TTS: A fast TTS architecture with conditional flow matching

Shivam Mehta, Ruibo Tu, Jonas Beskow, Éva Székely, Gustav Eje Henter
9/6/2023
eess.AScs.HCcs.LG

No official code found. Best implementation: shivammehta25/Matcha-TTS (1,350 stars, MIT, last commit 2026-04-27). 1 more below.

Abstract

We introduce Matcha-TTS, a new encoder-decoder architecture for speedy TTS acoustic modelling, trained using optimal-transport conditional flow matching (OT-CFM). This yields an ODE-based decoder capable of high output quality in fewer synthesis steps than models trained using score matching. Careful design choices additionally ensure each synthesis step is fast to run. The method is probabilistic, non-autoregressive, and learns to speak from scratch without external alignments. Compared to strong pre-trained baseline models, the Matcha-TTS system has the smallest memory footprint, rivals the speed of the fastest models on long utterances, and attains the highest mean opinion score in a listening test. Please see https://shivammehta25.github.io/Matcha-TTS/ for audio examples, code, and pre-trained models.

View on arXivView PDF

Code Implementations(2)

shivammehta25/Matcha-TTSCommunity100%
MIT

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

1,350215Shell, CythonSep 5, 20235 months agoMIT
deep-learningdiffusion-modeldiffusion-modelsflow-matchingmachine-learning+7 more
ayutaz/Matcha-TTS-JP-oldCommunity68%
MIT

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

00Shell, CythonOct 11, 20249 months agoMIT

Similar Papers

Shortcut Flow Matching for Speech Enhancement: Step-Invariant flows via single stage training

82% match2025

LiteVSR: Efficient Visual Speech Recognition by Learning from Speech Representations of Unlabeled Data

81% match2023

Audio-Native Speech Recognition with a Frozen Discrete-Diffusion Language Model

81% match2026

Qwen-Audio-Jev: A Unified Interface for Audio Decisions

81% match2026

Beyond U-Net: A Latent-Representation-Aligned Skip-Free Backbone for Flow-Matching Speech Enhancement

80% match2026

Quantize More, Lose Less: Autoregressive Generation from Residually Quantized Speech Representations

80% match2025

Cite this paper

@article{mehta2023matchatts,
  title  = {Matcha-TTS: A fast TTS architecture with conditional flow matching},
  author = {Shivam Mehta and Ruibo Tu and Jonas Beskow and Éva Székely and Gustav Eje Henter},
  year   = {2023},
  eprint = {2309.03199},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2309.03199v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2309.03199.svg)](https://codeofpaper.com/paper/2309.03199)

Discussion