| data: |
| block_size: 512 |
| cnhubertsoft_gate: 10 |
| duration: 1 |
| encoder: vec256l9 |
| encoder_hop_size: 320 |
| encoder_out_channels: 256 |
| encoder_sample_rate: 16000 |
| extensions: |
| - wav |
| sampling_rate: 44100 |
| training_files: filelists/train.txt |
| validation_files: filelists/val.txt |
| device: cuda |
| env: |
| expdir: logs/44k/diffusion |
| gpu_id: 0 |
| infer: |
| method: dpm-solver |
| speedup: 10 |
| model: |
| n_chans: 512 |
| n_hidden: 256 |
| n_layers: 20 |
| n_spk: 28 |
| type: Diffusion |
| use_pitch_aug: true |
| spk: |
| biaobei: 0 |
| buyizi: 1 |
| kirizu: 2 |
| m4_alto_1: 3 |
| m4_alto_2: 4 |
| m4_alto_3: 5 |
| m4_alto_4: 6 |
| m4_alto_5: 7 |
| m4_alto_6: 8 |
| m4_alto_7: 9 |
| m4_bass_1: 10 |
| m4_bass_2: 11 |
| m4_bass_3: 12 |
| m4_soprano_1: 13 |
| m4_soprano_2: 14 |
| m4_soprano_3: 15 |
| m4_tenor_1: 16 |
| m4_tenor_2: 17 |
| m4_tenor_3: 18 |
| m4_tenor_4: 19 |
| m4_tenor_5: 20 |
| m4_tenor_6: 21 |
| m4_tenor_7: 22 |
| nyaru: 23 |
| opencpop: 24 |
| paimon: 25 |
| sanyueqi: 26 |
| yousa: 27 |
| train: |
| amp_dtype: fp16 |
| batch_size: 48 |
| cache_all_data: false |
| cache_device: cpu |
| cache_fp16: true |
| decay_step: 100000 |
| epochs: 100000 |
| gamma: 0.5 |
| interval_force_save: 10000 |
| interval_log: 10 |
| interval_val: 2000 |
| lr: 0.0002 |
| num_workers: 2 |
| save_opt: false |
| weight_decay: 0 |
| vocoder: |
| ckpt: pretrain/nsf_hifigan/model |
| type: nsf-hifigan |
|
|
