// audio · marytts/marytts

marytts

marytts é um projeto de áudio de código aberto de marytts (marytts/marytts no GitHub). Tem 2.582 estrelas no GitHub e 771 forks, é publicado sob a licença NOASSERTION e foi atualizado pela última vez no GitHub em 17 de janeiro de 2025.

MaryTTS — open-source, multilingual text-to-speech platform written in pure Java

★ 2,6 mil estrelas⑂ 771 forksNOASSERTIONatualizado jan 2025

Ver no GitHub ↗Official site ↗

▦ visual: o README deste projeto não inclui nenhuma captura incorporável,
então aqui vai um marcador. O guia abaixo está completo mesmo assim.

// dados do projeto

Dados do projeto

Repositório: marytts/marytts
Estrelas no GitHub: 2.582
Forks: 771
Licença: NOASSERTION
Última atualização no GitHub: 17 de janeiro de 2025
Categoria: Áudio
Tipo de ficha: Guia de instalação completo
Site oficial: Official site

// como funciona

Como funciona

Clarify your goal
Decide whether you will run the standalone server (easiest from the releases page) or build from source / embed MaryTTS in a Java project.
Get MaryTTS
Download a ready-to-use package from the GitHub releases page, or clone the source repo.
Install and configure
From source, run it with Gradle; optionally start the voice installer GUI to download more voices, then restart the server.
Run your first synthesis
Open the server at http://localhost:59125, or send it an HTTP synthesis request from Python/shell using the linked txt2wav examples.

// recursos

Por que as pessoas gostam tanto

Pure Java, cross-platform

The whole client-server system is written in pure Java and runs wherever a JVM runs.

Multilingual platform

Designed as a multilingual TTS platform with downloadable voices; the default artifact referenced in the README is a US English HSMM voice.

HTTP server API

Any programming language can synthesize by compiling, starting the server and querying it over HTTP (examples for Python 3 and shell are linked).

Voice installer GUI

./gradlew runInstallerGui downloads and installs additional voices (restart the server to use them).

Embeddable via Maven/Gradle

Declare a dependency on MaryTTS voice artifacts to use it inside your own Java projects.

Distribution packaging

Gradle tasks (distZip/distTar/installDist) build a standalone server distribution, and user dictionaries can extend pronunciations.

// execute você mesmo

Execute você mesmo

Clone the repo
Clone MaryTTS (or download a ready package from the releases page instead).
```
git clone https://github.com/marytts/marytts.git
cd marytts
```
Build (optional)
Compile and test all modules and create outputs under build/.
```
./gradlew build
```
Run the server
Start a MaryTTS server and open it in your browser (port and address are configurable with -Dsocket.port/-Dsocket.addr).
```
./gradlew run  # http://localhost:59125
```

// casos de uso

O que você pode criar

A self-hosted TTS server for applications in several languages
Embedding speech synthesis in a Java product
Offline narration where cloud TTS is not an option
Research and customization of a classic statistical-parametric TTS platform

// perguntas frequentes

Perguntas frequentes

Who maintains MaryTTS?

The marytts open-source project maintains it; it originated from DFKI's MARY platform (older docs at mary.dfki.de and the marytts-wiki).

Is MaryTTS free?

Yes — the README states the code is under the LGPL version 3 license. (GitHub's license field shows NOASSERTION; the README/license file states LGPL-3.)

Does it need a paid API key?

No — synthesis runs locally on your own server; there is no cloud API involved.

How does it compare to pyttsx3 or gTTS?

pyttsx3 wraps your OS's voices in Python and gTTS calls Google Translate's endpoint. MaryTTS is a standalone multilingual server platform you host, with its own downloadable voices and an HTTP API for any language.

How do I get started?

Os outros projetos de Áudio com mais estrelas no catálogo do LibreAI — a lista realista de alternativas se marytts não for o ideal para você.

MoneyPrinterTurbo★ 128,3 mil
利用 AI 大模型和自动化工作流，根据主题或关键词一键生成高清短视频。Generate HD short videos from a topic or keyword with an automated AI workflow.
Whisper★ 109,9 mil
Robust Speech Recognition via Large-Scale Weak Supervision
GPT-SoVITS★ 62,3 mil
1 min voice data can also be used to train a good TTS model! (few shot voice cloning)
Real-Time-Voice-Cloning★ 60,2 mil
Clone a voice from 5 seconds of audio and generate arbitrary speech in real time
whisper.cpp★ 54,1 mil
Port of OpenAI's Whisper model in C/C++

// mais em audio

Mais em Áudio

Projetos de Áudio próximos no catálogo, os mais próximos em estrelas de marytts: continue explorando dos dois lados dele.

gTTS★ 2,6 mil
gTTS — Python library and CLI that turns text into spoken mp3 via Google Translate's speech API
pyttsx3★ 2,5 mil
Offline text-to-speech for Python — uses your operating system's own voices, no internet needed
VieNeu-TTS★ 2,8 mil
Vietnamese TTS with instant voice cloning — on-device, real-time CPU inference, 48 kHz audio
VoiceMem★ 2,4 mil
VoiceMem — streaming dual-brain memory for voice agents: a left brain for facts, a right brain for emotion
MARS5-TTS★ 2,8 mil
MARS5 speech model from CAMB.AI — clone a voice from 5 seconds of audio with striking prosody
hifi-gan★ 2,4 mil
GAN vocoder that turns mel-spectrograms into high-fidelity speech audio, fast enough for real-time use.

Baixe o marytts

Grátis e aberto: pegue o código-fonte, dê uma estrela no repositório e comece a criar.

Ver no GitHub ↗← Mais ferramentas de Áudio