libgdx-tts

Library for a text-to-speech synthesizer for LibGDX. Available for desktop and android apps.

Releases

For the latest release, download the jar here:

https://github.com/auraxangelic/libgdx-tts/releases

Usage

1. Download the jar from the releases.
2. Put the jar into your libGDX project.
3. Include the jar in your core module's build.gradle with something like the following:

dependencies {
    ...
    
    api fileTree(dir: "../libs", include: ["*.jar"])
}

4. Make sure you refresh gradle.
5. At any point after your Game class has been initialized, you my use the TextSpeech class to read any string.

import com.reikaxubia.libgdxtts.TextSpeech
...

    TextSpeech.speak("testing 1 2 3")
    // TextSpeech.INSTANCE.speak("testing 1 2 3") // in java

6. The text to speech will run asynchronously. If you want to block until it's finished speaking, you can run: TextSpeech.waitUntilFinished()/TextSpeech.INSTANCE.waitUntilFinished()
7. The TextSpeech class allocates and runs on its own thread. Make sure to deallocate the thread when your game exits by running TextSpeech.deallocate()/TextSpeech.INSTANCE.deallocate(), or run exitProcess(0)/System.exit(0).

Known Issues

Currently doesn't work for iOS robovm on libGDX version 1.12.1, but should be working once a new libGDX version is released. The PR to make it work for iOS is here: libgdx/libgdx#7371 You can have it work for iOS now by building the latest main branch by following steps here: https://libgdx.com/dev/from-source/

This library isn't uploaded to maven central yet, so you must download the binaries manualy or clone the source yourself to use it.

On desktop, when a new speak call is made, the audio will play a tiny bit of the end of the previous speak call. This is due to how the AudioDevice is currently implemented for desktop.

Information

Created using FreeTTS source: https://freetts.sourceforge.io/ by replacing its default audio player with the libGDX one Gdx.audio#newAudioDevice.