Deepfake Multimodal Weights
Trained weights for Multi-ModelDeepfakeDetection, a CLI that scores a video for manipulation by fusing three independent modalities (visual frames, audio track, spoken transcript).
Source du modèle
- Type
- Other
- Taille du fichier
- 774 KB–354 MB
- Modèle de base
- prithivMLmods/deepfake-detector-model-v1
- Version
- vClf ep6 torchscript
- Mis à jour
- 19 oct. 2026
- Source
- Hugging Face
Description de la source
Multi-Modal Deepfake Detection — trained weights
Trained weights for Multi-Model_Deepfake_Detection, a CLI that scores a video for manipulation by fusing three independent modalities (visual frames, audio track, spoken transcript).
Do not download these files by hand — clone the GitHub repo and run
python download_weights.py, which places every file where config.yaml
expects it.
Contents
| File | Role | Val metric |
|---|---|---|
video_model_colab_ft/ | SigLIP ViT, full fine-tune on FaceForensics++ (c23) face crops | AUC 0.861 / frame, 0.939 / video |
clf_ep6_torchscript.pt | MLP over WavLM-base-plus embeddings (WaveFake + JSUT) | AUC 0.999 |
text_model/head_best.pt | Linear head over DistilBERT for transcript claim scoring (FibVID) | AUC 0.909 |
fusion_model.pkl | HistGradientBoosting meta-classifier over the three probabilities | AUC 0.998 (simulated joint data) |
Validation methodology and known limitations (audio language confound, simulated fusion training data) are documented in the project README.
Licensing
Code and these trained heads: MIT. The video model is a fine-tune of prithivMLmods/deepfake-detector-model-v1 and retains its upstream terms. Training datasets are not redistributed here: FaceForensics++ requires a signed EULA; WaveFake is CC BY-SA; JSUT is for research use.
Sources
1 source
Sources
1 sourceVérifié 19 oct.
Download files
Artefacts du modèle
3 artefacts
Artefacts du modèle
3 artefactsvideo_model_colab_ft/model.safetensors
safetensors · 354 MB · SHA-256 9c93e5cb93c9…9c25 · Hugging Face
Download fileExtraits de sources
2 extraits
Extraits de sources
2 extraits--- license: mit base_model: prithivMLmods/deepfake-detector-model-v1 tags: - deepfake-detection - multimodal - video-classification - audio-classification --- # Multi-Modal Deepfake Detection — trained weights Trained weights for [Multi-Model_Deepfake_Detection](https://github.com/aua27/Multi-Model_Deepfake_Detection), a CLI that scores a video for manipulation by fusing three independent modalities (visual frames, audio track, spoken transcript). **Do not download these files by hand** — clone the GitHub repo and run `python download_weights.py`, which places every file where `config.yaml` expects it. ## Contents | File | Role | Val metric | |---|---|---| | `video_model_colab_ft/` | SigLIP ViT, full fine-tune on FaceForensics++ (c23) face crops | AUC 0.861 / frame, 0.939 / video | | `clf_ep6_torchscript.pt` | MLP over WavLM-base-plus embeddings (WaveFake + JSUT) | AUC 0.999 | | `text_model/head_best.pt` | Linear head over DistilBERT for transcript claim scoring (FibVID) | AUC 0.909 | | `fusion_model.pkl` | HistGradientBoosting meta-classifier over the three probabilities | AUC 0.998 (simulated joint data) | Validation methodology and known limitations (audio language confound, simulated fusion training data) are documented in the [project README](https://github.com/aua27/Multi-Model_Deepfake_Detection#known-limitations). ## Licensing Code and these trained heads: MIT. The video model is a fine-tune of [prithivMLmods/deepfake-detector-model-v1](https://huggingface.co/prithivMLmods/deepfake-detector-model-v1) and retains its upstream terms. Training datasets are **not** redistributed here: FaceForensics++ requires a signed EULA; WaveFake is CC BY-SA; JSUT is for research use.
Source context: 0 downloads · 0 likes · Pipeline audio-classification · Repo AUA27/deepfake-multimodal-weights
