Skip to content

This project implements TensorflowTTS in Tensorflow.js using Typescript, enabling real-time text-to-speech in the browser. With pre-trained model for English language, you can generate high-quality speech from text input.

License

Notifications You must be signed in to change notification settings

playerony/TensorFlowTTS-ts

Repository files navigation

TensorFlowTTS-ts

This project is a TypeScript implementation of TensorflowTTS's FastSpeech2 and MB-MelGAN LJSpeech models, enabling real-time text-to-speech in the browser for English language text.

Features

  • Real-time text-to-speech in the browser for English language text
  • Pre-trained models for English language
  • Based on TensorflowTTS's FastSpeech2 and MB-MelGAN LJSpeech models
  • Implementation in Tensorflow.js using TypeScript

Installation

yarn install

Usage

Clone the repository

git clone https://github.com/playerony/TensorFlowTTS-ts.git

Build an application

yarn build

Start an application

yarn open

Deployment

The application is deployed on Vercel and can be accessed at https://tensor-flow-tts-ts.vercel.app.

Credit

This implementation is based on the code from the following repository: https://github.com/MightyAlex200/tfjs-tts

License

This project is licensed under the MIT License - see the LICENSE file for details.

About

This project implements TensorflowTTS in Tensorflow.js using Typescript, enabling real-time text-to-speech in the browser. With pre-trained model for English language, you can generate high-quality speech from text input.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published