K_step: 1000
accumulate_grad_batches: 1
audio_num_mel_bins: 128
audio_sample_rate: 44100
binarization_args:
  shuffle: false
  with_align: true
  with_f0: true
  with_hubert: true
  with_spk_embed: false
  with_wav: false
binarizer_cls: preprocessing.SVCpre.SVCBinarizer
binary_data_dir: data/binary/aquapre
check_val_every_n_epoch: 10
choose_test_manually: false
clip_grad_norm: 1
config_path: F:\diff-svc-main\training\config_nsf.yaml
content_cond_steps: []
cwt_add_f0_loss: false
cwt_hidden_size: 128
cwt_layers: 2
cwt_loss: l1
cwt_std_scale: 0.8
datasets:
- opencpop
debug: false
dec_ffn_kernel_size: 9
dec_layers: 4
decay_steps: 20000
decoder_type: fft
dict_dir: ''
diff_decoder_type: wavenet
diff_loss_type: l2
dilation_cycle_length: 4
dropout: 0.1
ds_workers: 4
dur_enc_hidden_stride_kernel:
- 0,2,3
- 0,2,3
- 0,1,3
dur_loss: mse
dur_predictor_kernel: 3
dur_predictor_layers: 5
enc_ffn_kernel_size: 9
enc_layers: 4
encoder_K: 8
encoder_type: fft
endless_ds: false
f0_bin: 256
f0_max: 1100.0
f0_min: 40.0
f0_static: '{"28.0": 0.07, "29.0": 0.03, "31.0": 0.05, "32.0": 0.08, "33.0": 0.12,
  "34.0": 0.02, "35.0": 0.06, "36.0": 0.02, "37.0": 0.01, "38.0": 0.1, "39.0": 0.05,
  "40.0": 0.09, "41.0": 0.14, "42.0": 0.16, "43.0": 0.03, "44.0": 0.42, "45.0": 0.74,
  "46.0": 1.13, "47.0": 1.49, "48.0": 1.76, "49.0": 2.59, "50.0": 3.03, "51.0": 2.71,
  "52.0": 1.93, "53.0": 1.11, "54.0": 0.78, "55.0": 3.33, "56.0": 20.38, "57.0": 69.6,
  "58.0": 167.04, "59.0": 245.1, "60.0": 318.87, "61.0": 373.41, "62.0": 434.86, "63.0":
  415.63, "64.0": 448.97, "65.0": 452.99, "66.0": 474.88, "67.0": 471.54, "68.0":
  455.78, "69.0": 421.71, "70.0": 372.06, "71.0": 323.85, "72.0": 292.8, "73.0": 238.94,
  "74.0": 190.5, "75.0": 132.86, "76.0": 88.03, "77.0": 53.16, "78.0": 32.96, "79.0":
  23.66, "80.0": 14.74, "81.0": 8.54, "82.0": 5.0, "83.0": 3.32, "84.0": 2.29, "85.0":
  0.91, "total_time": 6576.43}'
ffn_act: gelu
ffn_padding: SAME
fft_size: 2048
fmax: 16000
fmin: 40
fs2_ckpt: ''
gaussian_start: true
gen_dir_name: ''
gen_tgt_spk_id: -1
hidden_size: 256
hop_size: 512
hubert_gpu: false
hubert_path: checkpoints/hubert/hubert_soft.pt
infer: false
keep_bins: 128
lambda_commit: 0.25
lambda_energy: 0.0
lambda_f0: 1.0
lambda_ph_dur: 0.3
lambda_sent_dur: 1.0
lambda_uv: 1.0
lambda_word_dur: 1.0
load_ckpt: ''
log_interval: 100
loud_norm: false
lr: 0.0008
max_beta: 0.02
max_epochs: 3000
max_eval_sentences: 1
max_eval_tokens: 60000
max_frames: 42000
max_input_tokens: 6000
max_sentences: 88
max_tokens: 128000
max_updates: 1000000
mel_loss: ssim:0.5|l1:0.5
mel_vmax: 1.5
mel_vmin: -6.0
min_level_db: -120
no_fs2: true
norm_type: gn
num_ckpt_keep: 10
num_heads: 2
num_sanity_val_steps: 1
num_spk: 1
num_test_samples: 0
num_valid_plots: 10
optimizer_adam_beta1: 0.9
optimizer_adam_beta2: 0.98
out_wav_norm: false
pe_ckpt: checkpoints/0102_xiaoma_pe/model_ckpt_steps_60000.ckpt
pe_enable: false
perform_enhance: true
pitch_ar: false
pitch_enc_hidden_stride_kernel:
- 0,2,5
- 0,2,5
- 0,2,5
pitch_extractor: parselmouth
pitch_loss: l2
pitch_norm: log
pitch_type: frame
pndm_speedup: 10
pre_align_args:
  allow_no_txt: false
  denoise: false
  forced_align: mfa
  txt_processor: zh_g2pM
  use_sox: true
  use_tone: false
pre_align_cls: data_gen.singing.pre_align.SingingPreAlign
predictor_dropout: 0.5
predictor_grad: 0.1
predictor_hidden: -1
predictor_kernel: 5
predictor_layers: 5
prenet_dropout: 0.5
prenet_hidden_size: 256
pretrain_fs_ckpt: ''
processed_data_dir: xxx
profile_infer: false
raw_data_dir: data/raw/aquapre
ref_norm_layer: bn
rel_pos: true
reset_phone_dict: true
residual_channels: 512
residual_layers: 20
save_best: false
save_ckpt: true
save_codes:
- configs
- modules
- src
- utils
save_f0: true
save_gt: false
schedule_type: linear
seed: 1234
sort_by_len: true
speaker_id: aqua
spec_max:
- 0.18377557396888733
- -0.33469653129577637
- -0.3073468506336212
- -0.21027648448944092
- 0.23178215324878693
- 0.5297451019287109
- 0.7021887898445129
- 0.7711099982261658
- 0.7912386059761047
- 0.6609739065170288
- 0.649876058101654
- 0.6327046751976013
- 0.6892049908638
- 0.6026111841201782
- 0.6834777593612671
- 0.7417489886283875
- 0.6040375828742981
- 0.5854794383049011
- 0.7123280167579651
- 0.5886657238006592
- 0.6135984063148499
- 0.5388530492782593
- 0.5932422280311584
- 0.535581111907959
- 0.57913738489151
- 0.6827316880226135
- 0.6265526413917542
- 0.6557696461677551
- 0.6586976647377014
- 0.5687282085418701
- 0.6218562722206116
- 0.6349128484725952
- 0.6176865100860596
- 0.6212958097457886
- 0.6277656555175781
- 0.5551338195800781
- 0.6126622557640076
- 0.5821346640586853
- 0.577056348323822
- 0.5649800300598145
- 0.5984634757041931
- 0.4873456656932831
- 0.47209471464157104
- 0.4387756586074829
- 0.4690910577774048
- 0.4616055190563202
- 0.3555675446987152
- 0.3898852467536926
- 0.3676068186759949
- 0.4632047414779663
- 0.37983986735343933
- 0.3877682685852051
- 0.3099276125431061
- 0.3261813223361969
- 0.34168118238449097
- 0.3004901111125946
- 0.3512653112411499
- 0.2647061347961426
- 0.2685043215751648
- 0.20390087366104126
- 0.1825377196073532
- 0.22067485749721527
- 0.20306138694286346
- 0.12710601091384888
- 0.10927848517894745
- 0.1117628887295723
- 0.14148156344890594
- 0.122605100274086
- 0.08032718300819397
- 0.12159623205661774
- -0.04923255369067192
- -0.07824847847223282
- 0.03441360592842102
- 0.07093964517116547
- -0.1269683688879013
- 0.0027632638812065125
- -0.045093610882759094
- -0.04115259647369385
- 0.029067598283290863
- -0.009453626349568367
- -0.0470033697783947
- -0.04894810542464256
- -0.06236470118165016
- -0.20086997747421265
- -0.2363593578338623
- -0.17289961874485016
- -0.219277486205101
- -0.2934815585613251
- -0.30551621317863464
- -0.2513120770454407
- -0.26792851090431213
- -0.33068278431892395
- -0.37532031536102295
- -0.365634560585022
- -0.3379015326499939
- -0.26979681849479675
- -0.20316314697265625
- -0.2109878957271576
- -0.16927000880241394
- -0.1698305308818817
- -0.2739156186580658
- -0.2700604200363159
- -0.32284122705459595
- -0.44529229402542114
- -0.4002469480037689
- -0.2441970407962799
- -0.19795942306518555
- -0.2462945580482483
- -0.0673084482550621
- -0.22117790579795837
- -0.21418607234954834
- -0.39467209577560425
- -0.4388139843940735
- -0.3227368891239166
- -0.30530503392219543
- -0.3201104998588562
- -0.39839836955070496
- -0.464596688747406
- -0.5399728417396545
- -0.5515261292457581
- -0.520453691482544
- -0.6714966893196106
- -0.6414765119552612
- -0.6108742356300354
- -0.6762520670890808
- -0.7067146301269531
- -0.7586700320243835
- -0.6640384793281555
spec_min:
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.999994277954102
- -4.989471912384033
- -4.999994277954102
spk_cond_steps: []
stop_token_weight: 5.0
task_cls: training.task.SVC_task.SVCTask
test_ids: []
test_input_dir: ''
test_num: 0
test_prefixes:
- test
test_set_name: test
timesteps: 1000
train_set_name: train
use_cn_hubert: false
use_crepe: true
use_denoise: false
use_energy_embed: false
use_gt_dur: false
use_gt_f0: false
use_midi: false
use_nsf: true
use_pitch_embed: true
use_pos_embed: true
use_spk_embed: false
use_spk_id: false
use_split_spk_id: false
use_uv: false
use_var_enc: false
use_vec: false
val_check_interval: 2000
valid_num: 0
valid_set_name: valid
vocoder: network.vocoders.nsf_hifigan.NsfHifiGAN
vocoder_ckpt: checkpoints/nsf_hifigan/model
warmup_updates: 2000
wav2spec_eps: 1e-6
weight_decay: 0
win_size: 2048
work_dir: checkpoints/aquapre