Download config.json from GeWu-Lab/SteerSound: direct link, hf CLI and curl.
- Browser
- Download file 1.23 kB
-
https://huggingface.co/GeWu-Lab/SteerSound/resolve/main/config.json
- Command line
-
hf download hf://GeWu-Lab/SteerSound/config.json
-
curl -L -o config.json https://huggingface.co/GeWu-Lab/SteerSound/resolve/main/config.json
1.23 kB
| { | |
| "format_version": 1, | |
| "model_type": "steersound", | |
| "dit": { | |
| "in_channels": 64, | |
| "out_channels": 64, | |
| "context_in_dim": 768, | |
| "hidden_size": 1536, | |
| "mlp_ratio": 4.0, | |
| "num_heads": 12, | |
| "depth": 2, | |
| "depth_single_blocks": 4, | |
| "axes_dim": [ | |
| 16, | |
| 56, | |
| 56 | |
| ], | |
| "theta": 10000, | |
| "qkv_bias": true, | |
| "guidance_embed": true, | |
| "patchify_proj_sharing_weight": true, | |
| "sharing_weight": true, | |
| "mean_txt": true, | |
| "enable_flash_attn": false, | |
| "vol_embed": true, | |
| "Pitch_embed": true | |
| }, | |
| "autoencoder": { | |
| "in_channels": 1, | |
| "ch": 256, | |
| "out_ch": 1, | |
| "ch_mult": [ | |
| 1, | |
| 2, | |
| 4, | |
| 4 | |
| ], | |
| "num_res_blocks": 4, | |
| "z_channels": 16, | |
| "enable_flash_attn": false | |
| }, | |
| "sample_rate": 16000, | |
| "duration_seconds": 10, | |
| "latent_channels": 16, | |
| "latent_height": 125, | |
| "latent_width": 8, | |
| "num_inference_steps": 50, | |
| "max_text_length": 128, | |
| "default_dtype": "bfloat16", | |
| "task_ids": { | |
| "add": 0, | |
| "remove": 1, | |
| "generate": 2, | |
| "replace": 3, | |
| "volume": 4, | |
| "pitch": 5 | |
| }, | |
| "change_volume_range_db": [ | |
| -5, | |
| 5 | |
| ], | |
| "change_pitch_range_semitones": [ | |
| -4, | |
| 4 | |
| ], | |
| "pitch_step_semitones": 0.5 | |
| } | |