/text-to-audio2face

Web interface to convert text to speech and route it to an Audio2Face streaming player.

Primary LanguageJavaScriptMIT LicenseMIT

Overview

text-to-Audio2Face is a Node.js application designed to convert text into speech and then stream the audio to the Audio2Face Streaming Audio Player for real-time facial animation.

Features

  • Toggle between manual mode, where you input text directly, and AI mode, which utilizes ChatGPT for generating text based on prompts.
  • Option to route the resulting audio stream directly to the browser, allowing users to listen to the converted text without using Audio2Face.
  • It supports real-time streaming of text generated by ChatGPT, ensuring immediate audio output for dynamic content.
  • The app integrates with Audio2Face, sending the generated TTS audio data to Audio2Face via gRPC to generate facial animations for digital avatars or characters.
  • Plans to support multiple TTS providers: Future updates promise the ability to switch between various text-to-speech providers, enhancing versatility and user choice. (Coming soon)

Note: This project is actively being developed, and additional features are planned for future releases.