Fix garbled/off-distribution outputs: align empty-subtitle prompt to reference get_prompt_for_multimodal template

Fix silently-dropped audio modality: harden ffmpeg fallback to use silence when a clip has no audio stream instead of dropping audio

Upload app.py with huggingface_hub

Upload app.py with huggingface_hub

Upload app.py with huggingface_hub

Upload app.py with huggingface_hub

Upload my_affectgpt/models/ImageBind/data.py with huggingface_hub

Upload my_affectgpt/models/ImageBind/data.py with huggingface_hub

Upload requirements.txt with huggingface_hub

Upload my_affectgpt/models/ImageBind/data.py with huggingface_hub

Upload app.py with huggingface_hub

Upload app.py with huggingface_hub

Upload folder using huggingface_hub

Upload app.py with huggingface_hub

Upload app.py with huggingface_hub

Upload requirements.txt with huggingface_hub

Upload requirements.txt with huggingface_hub

Upload README.md with huggingface_hub

Upload folder using huggingface_hub

initial commit