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/Unnamed
check_val_every_n_epoch: 10
choose_test_manually: false
clip_grad_norm: 1
config_path: 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
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: true
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: pretrain/nehito_ckpt_steps_1000000.ckpt
log_interval: 100
loud_norm: false
lr: 5.0e-05
max_beta: 0.02
max_epochs: 3000
max_eval_sentences: 1
max_eval_tokens: 60000
max_frames: 42000
max_input_tokens: 60000
max_sentences: 12
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/Unnamed
ref_norm_layer: bn
rel_pos: true
reset_phone_dict: true
residual_channels: 384
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: Unnamed
spec_max:
- -0.4884430170059204
- 0.004534448496997356
- 0.5684943795204163
- 0.6527385115623474
- 0.659079372882843
- 0.7416915893554688
- 0.844637930393219
- 0.806076169013977
- 0.7238750457763672
- 0.9744535088539124
- 0.9476388692855835
- 0.9883336424827576
- 1.0821290016174316
- 1.046391248703003
- 0.9829667806625366
- 1.0163493156433105
- 0.9825412631034851
- 0.9834834337234497
- 0.9811502695083618
- 1.128888726234436
- 1.186057209968567
- 1.112004280090332
- 1.1282787322998047
- 1.051572322845459
- 1.0510444641113281
- 1.0110565423965454
- 0.9236567616462708
- 0.8036720156669617
- 0.8383486270904541
- 0.7735869288444519
- 0.9303848743438721
- 1.1257890462875366
- 1.1610286235809326
- 1.0335885286331177
- 1.0645352602005005
- 1.0619306564331055
- 1.1310148239135742
- 1.1191954612731934
- 1.1307402849197388
- 0.8837698698043823
- 1.1153966188430786
- 1.1045044660568237
- 1.0479614734649658
- 0.9491603374481201
- 0.9858523011207581
- 0.9226155281066895
- 0.9469702839851379
- 0.8791896104812622
- 0.997624933719635
- 0.9068642854690552
- 0.9575618505477905
- 0.8551340699195862
- 0.8397778272628784
- 0.8908605575561523
- 0.7986546158790588
- 0.7983465194702148
- 0.6965265274047852
- 0.640673041343689
- 0.6690735220909119
- 0.5631484985351562
- 0.48587048053741455
- 0.5326520800590515
- 0.4286036193370819
- 0.35252484679222107
- 0.3290073573589325
- 0.4754445552825928
- 0.3632410168647766
- 0.391481876373291
- 0.20288512110710144
- 0.18305960297584534
- 0.1539602279663086
- 0.03451670706272125
- -0.16881510615348816
- -0.02030198462307453
- 0.10024689882993698
- -0.023952053859829903
- 0.05635542422533035
- 0.10877621918916702
- 0.006155031267553568
- 0.07318088412284851
- 0.14075303077697754
- 0.057870157063007355
- -0.0520513579249382
- 0.1741427332162857
- -0.11464552581310272
- 0.03305494412779808
- -0.06897418200969696
- -0.12598733603954315
- -0.09894973039627075
- -0.2817802429199219
- -0.0825519785284996
- -0.3040400445461273
- -0.4998124837875366
- -0.36957985162734985
- -0.5409602522850037
- -0.49879470467567444
- -0.713716983795166
- -0.6545754671096802
- -0.6425778865814209
- -0.6178902387619019
- -0.47356730699539185
- -0.6165243983268738
- -0.5841533541679382
- -0.5759448409080505
- -0.5498068332672119
- -0.4661938548088074
- -0.5811225771903992
- -0.614664614200592
- -0.3902229070663452
- -0.7037366032600403
- -0.7260795831680298
- -0.7540019750595093
- -0.8360528945922852
- -0.8374698758125305
- -0.8328713178634644
- -0.9081047177314758
- -0.9679695963859558
- -0.9587443470954895
- -1.0706337690353394
- -0.9818469285964966
- -0.8360191583633423
- -0.9938981533050537
- -1.0823708772659302
- -1.0617167949676514
- -1.1093820333480835
- -1.1300138235092163
- -1.2141350507736206
- -1.3147293329238892
spec_min:
- -4.473258972167969
- -4.244492530822754
- -4.390527725219727
- -4.209497928619385
- -4.446024417877197
- -4.3960185050964355
- -4.164802551269531
- -4.5063300132751465
- -4.608232021331787
- -4.251623630523682
- -4.4799604415893555
- -4.733210563659668
- -4.411860466003418
- -4.609100818634033
- -4.726972579956055
- -4.428761959075928
- -4.487612247467041
- -4.525552749633789
- -4.480506896972656
- -4.589383125305176
- -4.608384132385254
- -4.385376453399658
- -4.816161632537842
- -4.8706955909729
- -4.848956108093262
- -4.431278705596924
- -4.999994277954102
- -4.818373203277588
- -4.527368068695068
- -4.872085094451904
- -4.894851207733154
- -4.511948585510254
- -4.534575939178467
- -4.57792854309082
- -4.444681644439697
- -4.628803253173828
- -4.74341344833374
- -4.85427713394165
- -4.723776817321777
- -4.7166008949279785
- -4.749168395996094
- -4.67240047454834
- -4.590690612792969
- -4.576009750366211
- -4.542308330535889
- -4.890907287597656
- -4.613001823425293
- -4.494126796722412
- -4.474257946014404
- -4.574635028839111
- -4.4817585945129395
- -4.651009559631348
- -4.478254795074463
- -4.523812770843506
- -4.546536922454834
- -4.535660266876221
- -4.470296859741211
- -4.577486991882324
- -4.541748046875
- -4.428532123565674
- -4.461862564086914
- -4.489077091217041
- -4.515830039978027
- -4.395663738250732
- -4.439975738525391
- -4.4290876388549805
- -4.397741794586182
- -4.478252410888672
- -4.399686336517334
- -4.45617151260376
- -4.434477806091309
- -4.442898750305176
- -4.5840277671813965
- -4.537542819976807
- -4.492046356201172
- -4.534677505493164
- -4.477104187011719
- -4.511618614196777
- -4.387601375579834
- -4.499236106872559
- -4.3717169761657715
- -4.4242024421691895
- -4.4055657386779785
- -4.429355144500732
- -4.4636993408203125
- -4.508528232574463
- -4.515079498291016
- -4.426190376281738
- -4.433525085449219
- -4.4200215339660645
- -4.421280860900879
- -4.400143623352051
- -4.419166088104248
- -4.429825305938721
- -4.436781406402588
- -4.51550817489624
- -4.518474578857422
- -4.495880603790283
- -4.483924865722656
- -4.409562587738037
- -4.3811845779418945
- -4.411908149719238
- -4.427165985107422
- -4.396549701690674
- -4.340637683868408
- -4.405435085296631
- -4.367630481719971
- -4.419083595275879
- -4.389026165008545
- -4.371067047119141
- -4.370710372924805
- -4.3755269050598145
- -4.39500093460083
- -4.451773166656494
- -4.365351676940918
- -4.348028182983398
- -4.408270359039307
- -4.390385627746582
- -4.347931861877441
- -4.378237247467041
- -4.426717758178711
- -4.364233493804932
- -4.371546745300293
- -4.402477264404297
- -4.430750846862793
- -4.404538154602051
- -4.384459018707275
- -4.401677131652832
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_crepe: false
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: 1000
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/HokoHifi