add files

README.md

@@ -1,3 +1,17 @@
+Model Arch: TokSing
+
+Dataste: Opencpop
+
+You can run this model with espnet/svs2.
+
+```sh
+./run.sh --stage 8 \
+--kmeans_feature "multi/hubert_large_6+wavlm_large_6+wavlm_large_23" \
+--multi_token "hubert_large_ll60k_128_6 wavlm_large_128_6 wavlm_large_128_23" \
+--inference_model 300epoch.pth \
+--vocoder_file {vocoder_path}
+```
+
 ---
 license: cc-by-4.0
 ---

exp/svs_stats_raw_phn_none_zh/train/feats_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55ee5850cc8928c2849db7fd96af9269d1b6f6bf6e781192c566733deefe0910
+size 1402

exp/svs_stats_raw_phn_none_zh/train/pitch_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af7755bf050318d1edf0bce559fc7f019c925d6376a9a1101d69348d0cea725f
+size 770

exp/svs_train_toksing_raw_phn_none_zh/300epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f094124cc67c1110a4b24b86d1b5252c97fcc96de9d2e28bb00e88c5ad9f390d
+size 281020670

exp/svs_train_toksing_raw_phn_none_zh/config.yaml

@@ -0,0 +1,315 @@
+config: conf/tuning/train_toksing.yaml
+print_config: false
+log_level: INFO
+drop_last_iter: false
+dry_run: false
+iterator_type: sequence
+valid_iterator_type: null
+output_dir: exp/svs_train_toksing_raw_phn_none_zh
+ngpu: 1
+seed: 0
+num_workers: 10
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: null
+dist_rank: null
+local_rank: 0
+dist_master_addr: null
+dist_master_port: null
+dist_launcher: null
+multiprocessing_distributed: false
+unused_parameters: false
+sharded_ddp: false
+use_deepspeed: false
+deepspeed_config: null
+gradient_as_bucket_view: true
+ddp_comm_hook: null
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+use_tf32: false
+collect_stats: false
+write_collected_feats: false
+max_epoch: 300
+patience: null
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+-   - valid
+    - loss
+    - min
+-   - train
+    - loss
+    - min
+keep_nbest_models: 5
+nbest_averaging_interval: 0
+grad_clip: 1.0
+grad_clip_type: 2.0
+grad_noise: false
+accum_grad: 1
+no_forward_run: false
+resume: true
+train_dtype: float32
+use_amp: false
+log_interval: null
+use_matplotlib: true
+use_tensorboard: true
+create_graph_in_tensorboard: false
+use_wandb: false
+wandb_project: null
+wandb_id: null
+wandb_entity: null
+wandb_name: null
+wandb_model_log_interval: -1
+detect_anomaly: false
+use_adapter: false
+adapter: lora
+save_strategy: all
+adapter_conf: {}
+pretrain_path: null
+init_param: []
+ignore_init_mismatch: false
+freeze_param: []
+num_iters_per_epoch: null
+batch_size: 32
+valid_batch_size: null
+batch_bins: 1000000
+valid_batch_bins: null
+category_sample_size: 10
+train_shape_file:
+- exp/svs_stats_raw_phn_none_zh/train/text_shape.phn
+- exp/svs_stats_raw_phn_none_zh/train/singing_shape
+valid_shape_file:
+- exp/svs_stats_raw_phn_none_zh/valid/text_shape.phn
+- exp/svs_stats_raw_phn_none_zh/valid/singing_shape
+batch_type: sorted
+valid_batch_type: null
+fold_length:
+- 150
+- 256000
+sort_in_batch: descending
+shuffle_within_batch: false
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+chunk_excluded_key_prefixes: []
+chunk_default_fs: null
+chunk_max_abs_length: null
+chunk_discard_short_samples: true
+train_data_path_and_name_and_type:
+-   - dump/raw/tr_no_dev/text
+    - text
+    - text
+-   - dump/raw/tr_no_dev/wav.scp
+    - singing
+    - sound
+-   - dump/raw/tr_no_dev/label
+    - label
+    - duration
+-   - dump/raw/tr_no_dev/score.scp
+    - score
+    - score
+-   - dump/raw/tr_no_dev/token_multi_hubert_large_6+wavlm_large_6+wavlm_large_23
+    - discrete_token
+    - text_int
+valid_data_path_and_name_and_type:
+-   - dump/raw/dev/text
+    - text
+    - text
+-   - dump/raw/dev/wav.scp
+    - singing
+    - sound
+-   - dump/raw/dev/label
+    - label
+    - duration
+-   - dump/raw/dev/score.scp
+    - score
+    - score
+-   - dump/raw/dev/token_multi_hubert_large_6+wavlm_large_6+wavlm_large_23
+    - discrete_token
+    - text_int
+multi_task_dataset: false
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+allow_multi_rates: false
+valid_max_cache_size: null
+exclude_weight_decay: false
+exclude_weight_decay_conf: {}
+optim: adam
+optim_conf:
+    lr: 0.001
+    eps: 1.0e-06
+    weight_decay: 0.0
+scheduler: null
+scheduler_conf: {}
+token_list:
+- <blank>
+- <unk>
+- SP
+- i
+- AP
+- e
+- y
+- d
+- w
+- sh
+- ai
+- n
+- x
+- j
+- ian
+- u
+- l
+- h
+- b
+- o
+- zh
+- an
+- ou
+- m
+- q
+- z
+- en
+- g
+- ing
+- ei
+- ao
+- ang
+- uo
+- eng
+- t
+- a
+- ong
+- ui
+- k
+- f
+- r
+- iang
+- ch
+- v
+- in
+- iao
+- ie
+- iu
+- c
+- s
+- van
+- p
+- ve
+- uan
+- uang
+- ia
+- ua
+- uai
+- un
+- er
+- vn
+- iong
+- <sos/eos>
+odim: null
+model_conf: {}
+use_preprocessor: true
+token_type: phn
+bpemodel: null
+non_linguistic_symbols: null
+cleaner: null
+g2p: null
+fs: 16000
+discrete_token_layers: 3
+nclusters: 128
+score_feats_extract: syllable_score_feats
+score_feats_extract_conf:
+    fs: 16000
+    n_fft: 2048
+    win_length: 1280
+    hop_length: 320
+feats_extract: fbank
+feats_extract_conf:
+    n_fft: 2048
+    hop_length: 320
+    win_length: 1280
+    fs: 16000
+    fmin: 80
+    fmax: 7600
+    n_mels: 80
+normalize: global_mvn
+normalize_conf:
+    stats_file: exp/svs_stats_raw_phn_none_zh/train/feats_stats.npz
+svs: toksing
+svs_conf:
+    midi_dim: 129
+    duration_dim: 512
+    adim: 384
+    aheads: 4
+    elayers: 6
+    eunits: 1536
+    dlayers: 6
+    dunits: 1536
+    postnet_layers: 0
+    postnet_chans: 512
+    postnet_filts: 5
+    postnet_dropout_rate: 0.5
+    use_batch_norm: true
+    reduction_factor: 1
+    global_channels: -1
+    text_encoder_attention_heads: 2
+    text_encoder_ffn_expand: 4
+    text_encoder_blocks: 6
+    text_encoder_positionwise_layer_type: conv1d
+    text_encoder_positionwise_conv_kernel_size: 3
+    text_encoder_positional_encoding_layer_type: rel_pos
+    text_encoder_self_attention_layer_type: rel_selfattn
+    text_encoder_activation_type: swish
+    text_encoder_normalize_before: true
+    text_encoder_dropout_rate: 0.1
+    text_encoder_positional_dropout_rate: 0.0
+    text_encoder_attention_dropout_rate: 0.1
+    use_macaron_style_in_text_encoder: true
+    use_conformer_conv_in_text_encoder: false
+    text_encoder_conformer_kernel_size: -1
+    init_type: pytorch
+    use_masking: true
+    loss_function: FastSpeech1
+    loss_type: L1
+    lambda_out: 1
+    lambda_dur: 1
+    lambda_pitch: 1
+    lambda_vuv: 0.01
+    use_discrete_token: true
+    predict_pitch: true
+    codec_codebook: 0
+pitch_extract: dio
+pitch_extract_conf:
+    use_token_averaged_f0: false
+    use_log_f0: true
+    fs: 16000
+    n_fft: 2048
+    hop_length: 320
+    f0max: 800
+    f0min: 80
+    reduction_factor: 1
+pitch_normalize: null
+pitch_normalize_conf:
+    stats_file: exp/svs_stats_raw_phn_none_zh/train/pitch_stats.npz
+ying_extract: null
+ying_extract_conf: {}
+energy_extract: null
+energy_extract_conf: {}
+energy_normalize: null
+energy_normalize_conf: {}
+model_type: discrete_svs
+model_type_conf: {}
+required:
+- output_dir
+- token_list
+version: '202503'
+distributed: false

meta.yaml

@@ -0,0 +1,8 @@
+espnet: '202503'
+files:
+  model_file: exp/svs_train_toksing_raw_phn_none_zh/300epoch.pth
+python: 3.10.16 (main, Dec 11 2024, 16:24:50) [GCC 11.2.0]
+timestamp: 1750092456.204741
+torch: 2.5.1+cu124
+yaml_files:
+  train_config: exp/svs_train_toksing_raw_phn_none_zh/config.yaml

vocoder/checkpoint-250000steps.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53f74e4ccfa1a404a7d8661f1dade29f1a12fa86af5f215280f7664a9e26f867
+size 1091678223

vocoder/config.yml

@@ -0,0 +1,205 @@
+allow_cache: true
+batch_max_steps: 10240
+batch_size: 16
+config: conf/hifigan_token_16k_nodp_f0.v1.yaml
+dev_dumpdir: dump_token/dump_hb_6+wl_6+wl_23_f0/dev/raw
+dev_feats_scp: null
+dev_segments: null
+dev_wav_scp: null
+discriminator_adv_loss_params:
+  average_by_discriminators: false
+discriminator_grad_norm: -1
+discriminator_optimizer_params:
+  betas:
+  - 0.5
+  - 0.9
+  lr: 0.0002
+  weight_decay: 0.0
+discriminator_optimizer_type: Adam
+discriminator_params:
+  follow_official_norm: true
+  period_discriminator_params:
+    bias: true
+    channels: 32
+    downsample_scales:
+    - 3
+    - 3
+    - 3
+    - 3
+    - 1
+    in_channels: 1
+    kernel_sizes:
+    - 5
+    - 3
+    max_downsample_channels: 1024
+    nonlinear_activation: LeakyReLU
+    nonlinear_activation_params:
+      negative_slope: 0.1
+    out_channels: 1
+    use_spectral_norm: false
+    use_weight_norm: true
+  periods:
+  - 2
+  - 3
+  - 5
+  - 7
+  - 11
+  scale_discriminator_params:
+    bias: true
+    channels: 128
+    downsample_scales:
+    - 4
+    - 4
+    - 4
+    - 4
+    - 1
+    in_channels: 1
+    kernel_sizes:
+    - 15
+    - 41
+    - 5
+    - 3
+    max_downsample_channels: 1024
+    max_groups: 16
+    nonlinear_activation: LeakyReLU
+    nonlinear_activation_params:
+      negative_slope: 0.1
+    out_channels: 1
+  scale_downsample_pooling: AvgPool1d
+  scale_downsample_pooling_params:
+    kernel_size: 4
+    padding: 2
+    stride: 2
+  scales: 3
+discriminator_scheduler_params:
+  gamma: 0.5
+  milestones:
+  - 200000
+  - 400000
+  - 600000
+  - 800000
+discriminator_scheduler_type: MultiStepLR
+discriminator_train_start_steps: 0
+discriminator_type: HiFiGANMultiScaleMultiPeriodDiscriminator
+distributed: false
+duration_loss_params:
+  offset: 1.0
+  reduction: mean
+eval_interval_steps: 10000
+feat_match_loss_params:
+  average_by_discriminators: false
+  average_by_layers: false
+  include_final_outputs: true
+fft_size: null
+fmax: null
+fmin: null
+format: hdf5
+generator_adv_loss_params:
+  average_by_discriminators: false
+generator_grad_norm: -1
+generator_optimizer_params:
+  betas:
+  - 0.5
+  - 0.9
+  lr: 0.0002
+  weight_decay: 0.0
+generator_optimizer_type: Adam
+generator_params:
+  bias: true
+  channels: 512
+  in_channels: 768
+  kernel_size: 7
+  layer_num: 3
+  linear_channel: 256
+  nonlinear_activation: LeakyReLU
+  nonlinear_activation_params:
+    negative_slope: 0.1
+  num_embs: 1025
+  num_spk_embs: 0
+  out_channels: 1
+  resblock_dilations:
+  - - 1
+    - 3
+    - 5
+  - - 1
+    - 3
+    - 5
+  - - 1
+    - 3
+    - 5
+  resblock_kernel_sizes:
+  - 3
+  - 7
+  - 11
+  upsample_kernel_sizes:
+  - 20
+  - 16
+  - 4
+  - 4
+  upsample_scales:
+  - 10
+  - 8
+  - 2
+  - 2
+  use_additional_convs: true
+  use_embedding_feats: false
+  use_f0: true
+  use_fix_weight: false
+  use_weight_norm: true
+  use_weight_sum: true
+generator_scheduler_params:
+  gamma: 0.5
+  milestones:
+  - 200000
+  - 400000
+  - 600000
+  - 800000
+generator_scheduler_type: MultiStepLR
+generator_train_start_steps: 1
+generator_type: DiscreteSymbolF0Generator
+global_gain_scale: 1.0
+hop_size: 320
+lambda_adv: 1.0
+lambda_aux: 45.0
+lambda_feat_match: 2.0
+log_interval_steps: 100
+mel_loss_params:
+  fft_size: 1024
+  fmax: 8000
+  fmin: 0
+  fs: 16000
+  hop_size: 256
+  log_base: null
+  num_mels: 80
+  win_length: null
+  window: hann
+num_mels: 1
+num_save_intermediate_results: 4
+num_workers: 0
+outdir: exp/hb_6+wl_6+wl_23_f0/train_opencpop_hifigan_token_16k_nodp_f0.v1
+pin_memory: true
+pretrain: ''
+rank: 0
+remove_short_samples: false
+resume: exp/hb_6+wl_6+wl_23_f0/train_opencpop_hifigan_token_16k_nodp_f0.v1/checkpoint-27steps.pkl
+sampling_rate: 16000
+save_interval_steps: 50000
+train_dumpdir: dump_token/dump_hb_6+wl_6+wl_23_f0/train/raw
+train_feats_scp: null
+train_max_steps: 250000
+train_segments: null
+train_wav_scp: null
+trim_frame_size: 1024
+trim_hop_size: 256
+trim_silence: false
+trim_threshold_in_db: 20
+use_duration_loss: true
+use_f0: true
+use_feat_match_loss: true
+use_mel_loss: true
+use_multi_resolution_token: false
+use_stft_loss: false
+verbose: 1
+version: 0.6.2a
+win_length: null
+window: null