/kalita-server

A Java speech synthesizer and backend server for Kalita, a text-to-speech software with a focus on data minimization and user privacy.

Primary LanguageJavaMIT LicenseMIT

Logo Kalita

kalita-server

Kalita a is text-to-speech software with a special focus on data minimization and user privacy. We do not collect any personal data, do not set tracking cookies and do not outsource our service to third-party cloud solutions. The speech synthesis takes place on-premises on your own server and still offers many of the conveniences of a conventional readspeaker.

  • kalita-server is a server written in Python that provides the speech synthesis.
  • kalita-js is a JavaScript client for integration into a website that provides a graphical user interface.

🔗 Links and Resources

Type Links
💾 Installation TTS/README.md

Install TTS

Kalita is tested on Windows 10 with python 3.8.9.

Prequesites:

The underlying coquiTTS uses espeak-ng to convert graphemes to phonemes. You might need to install separately HERE. Or on Linux distributions via console:

> sudo apt-get install espeak-ng

To convert text to speech we need to install coquiTTS as initial setup:

> pip install TTS

Build Java Server

Continue with "Using the server" if you already have the .jar file.

> gradle wrapper
> gradlew shadowJar

The .jar file gets saved to ./build/libs/.

Using the server

Configure the address and port of the Java server and Python server in the config.properties file.

Start both servers:

> java -jar build/libs/kalita-server-java-1.0.jar
> python server.py

After the server started you can send requests to the server.

> GET http://localhost:8080/speak?text=Here%20comes%20your%20text