| { | |
| "architectures": [ | |
| "DiscoAudioModel" | |
| ], | |
| "audio_f": 128, | |
| "audio_fps": 16000, | |
| "audio_norm": false, | |
| "audio_rep": "wave16k", | |
| "audio_sr": 16000, | |
| "body_dims": 78, | |
| "class_name": "DiscoAudioModel", | |
| "dropout_prob": 0.1, | |
| "hands_dims": 180, | |
| "hidden_size": 512, | |
| "joint_mask": "local_upper", | |
| "model_type": "disco_audio", | |
| "motion_f": 256, | |
| "n_layer": 4, | |
| "name_pyfile": "models.disco_audio.modeling_disco_audio", | |
| "pose_dims": 258, | |
| "pose_fps": 15, | |
| "pose_rep": "smplx", | |
| "seed_frames": 4, | |
| "speaker_dims": 1, | |
| "speaker_f": 16, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.30.2" | |
| } | |