| { | |
| "_class_name": "Decoder", | |
| "_diffusers_version": "0.36.0", | |
| "attn_resolutions": [], | |
| "attn_type": "vanilla", | |
| "causality_axis": "height", | |
| "ch": 128, | |
| "ch_mult": [ | |
| 1, | |
| 2, | |
| 4 | |
| ], | |
| "dropout": 0.0, | |
| "give_pre_end": false, | |
| "is_causal": true, | |
| "mel_bins": 64, | |
| "mel_hop_length": 160, | |
| "mid_block_add_attention": false, | |
| "norm_type": "pixel", | |
| "num_res_blocks": 2, | |
| "out_ch": 2, | |
| "resamp_with_conv": true, | |
| "resolution": 256, | |
| "sample_rate": 16000, | |
| "tanh_out": false, | |
| "temb_ch": 0, | |
| "z_channels": 8 | |
| } | |