文件列表:

api (0, 2021-09-26)
api\Dockerfile (306, 2021-09-26)
api\Wav2Lip (0, 2021-09-26)
api\Wav2Lip\__init__.py (0, 2021-09-26)
api\Wav2Lip\audio.py (4507, 2021-09-26)
api\Wav2Lip\checkpoints (0, 2021-09-26)
api\Wav2Lip\color_syncnet_train.py (8771, 2021-09-26)
api\Wav2Lip\evaluation (0, 2021-09-26)
api\Wav2Lip\evaluation\gen_videos_from_filelist.py (7204, 2021-09-26)
api\Wav2Lip\evaluation\real_videos_inference.py (9173, 2021-09-26)
api\Wav2Lip\evaluation\scores_LSE (0, 2021-09-26)
api\Wav2Lip\evaluation\scores_LSE\SyncNetInstance_calc_scores.py (6801, 2021-09-26)
api\Wav2Lip\evaluation\scores_LSE\calculate_scores_LRS.py (1650, 2021-09-26)
api\Wav2Lip\evaluation\scores_LSE\calculate_scores_real_videos.py (1615, 2021-09-26)
api\Wav2Lip\evaluation\scores_LSE\calculate_scores_real_videos.sh (294, 2021-09-26)
api\Wav2Lip\evaluation\test_filelists (0, 2021-09-26)
api\Wav2Lip\evaluation\test_filelists\ReSyncED (0, 2021-09-26)
api\Wav2Lip\evaluation\test_filelists\ReSyncED\random_pairs.txt (3892, 2021-09-26)
api\Wav2Lip\evaluation\test_filelists\ReSyncED\tts_pairs.txt (515, 2021-09-26)
api\Wav2Lip\face_detection (0, 2021-09-26)
api\Wav2Lip\face_detection\__init__.py (183, 2021-09-26)
api\Wav2Lip\face_detection\api.py (2275, 2021-09-26)
api\Wav2Lip\face_detection\detection (0, 2021-09-26)
api\Wav2Lip\face_detection\detection\__init__.py (30, 2021-09-26)
api\Wav2Lip\face_detection\detection\core.py (4868, 2021-09-26)
api\Wav2Lip\face_detection\detection\sfd (0, 2021-09-26)
api\Wav2Lip\face_detection\detection\sfd\__init__.py (53, 2021-09-26)
api\Wav2Lip\face_detection\detection\sfd\bbox.py (4279, 2021-09-26)
api\Wav2Lip\face_detection\detection\sfd\detect.py (3769, 2021-09-26)
api\Wav2Lip\face_detection\detection\sfd\net_s3fd.py (5291, 2021-09-26)
api\Wav2Lip\face_detection\detection\sfd\sfd_detector.py (1809, 2021-09-26)
... ...

# 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 - [x] Text Summarization - [ ] Text Translation - [x] Speech Synthesis - [x] Video Generation - [x] Lip Synchronization - [ ] Multiple Language Support - [ ] Make Your Own Speaker ## How does it work We use [HuggingFace Transformers](https://huggingface.co/transformers/index.html) for Natural Language Understanding tasks and then generate audio with [NVIDIA's tacotron2](https://github.com/NVIDIA/tacotron2). We use the audio on predefined video material and adjust the speakers' lips using [Wav2Lip](https://github.com/Rudrabha/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`

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：