Skip to content

Latest commit

 

History

History
22 lines (15 loc) · 919 Bytes

README.md

File metadata and controls

22 lines (15 loc) · 919 Bytes

DeepFakeNews

Generate realistic news videos solely by providing your transcript. We summarize your news and generate a video of a speaker announcing your news.

Features

  • Text Summarization
  • Text Translation
  • Speech Synthesis
  • Video Generation
  • Lip Synchronization
  • Multiple Language Support
  • Make Your Own Speaker

How does it work

We use HuggingFace Transformers for Natural Language Understanding tasks and then generate audio with NVIDIA's tacotron2. We use the audio on predefined video material and adjust the speakers' lips using Wav2Lip.

The processing pipeline is integrated in FastAPI and gets accessed by a small React frontend. The entire application can be started using docker-compose.

Run the application

docker-compose up