| N: '' | |
| T: 1000 | |
| accumulate_grad_batches: 1 | |
| amp: false | |
| audio_channels: 1 | |
| audio_num_mel_bins: 80 | |
| audio_sample_rate: 22050 | |
| aux_context_window: 0 | |
| beta_0: 1.0e-06 | |
| beta_T: 0.01 | |
| binarization_args: | |
| reset_phone_dict: true | |
| reset_word_dict: true | |
| shuffle: false | |
| trim_eos_bos: false | |
| with_align: false | |
| with_f0: false | |
| with_f0cwt: false | |
| with_linear: false | |
| with_spk_embed: false | |
| with_spk_id: true | |
| with_txt: false | |
| with_wav: true | |
| with_word: false | |
| binarizer_cls: data_gen.tts.vocoder_binarizer.VocoderBinarizer | |
| binary_data_dir: data/binary/LJSpeech | |
| check_val_every_n_epoch: 10 | |
| clip_grad_norm: 1 | |
| clip_grad_value: 0 | |
| cond_channels: 80 | |
| debug: false | |
| dec_ffn_kernel_size: 9 | |
| dec_layers: 4 | |
| dict_dir: '' | |
| diffusion_step_embed_dim_in: 128 | |
| diffusion_step_embed_dim_mid: 512 | |
| diffusion_step_embed_dim_out: 512 | |
| disc_start_steps: 40000 | |
| discriminator_grad_norm: 1 | |
| dropout: 0.0 | |
| ds_workers: 1 | |
| enc_ffn_kernel_size: 9 | |
| enc_layers: 4 | |
| endless_ds: true | |
| eval_max_batches: -1 | |
| ffn_act: gelu | |
| ffn_padding: SAME | |
| fft_size: 1024 | |
| fmax: 7600 | |
| fmin: 80 | |
| frames_multiple: 1 | |
| gen_dir_name: '' | |
| generator_grad_norm: 10 | |
| griffin_lim_iters: 60 | |
| hidden_size: 256 | |
| hop_size: 256 | |
| infer: false | |
| inner_channels: 32 | |
| kpnet_conv_size: 3 | |
| kpnet_hidden_channels: 64 | |
| load_ckpt: '' | |
| loud_norm: false | |
| lr: 2e-4 | |
| lvc_kernel_size: 3 | |
| lvc_layers_each_block: 4 | |
| max_epochs: 1000 | |
| max_frames: 1548 | |
| max_input_tokens: 1550 | |
| max_samples: 25600 | |
| max_sentences: 20 | |
| max_tokens: 30000 | |
| max_updates: 1000000 | |
| max_valid_sentences: 1 | |
| max_valid_tokens: 60000 | |
| mel_loss: l1 | |
| mel_vmax: 1.5 | |
| mel_vmin: -6 | |
| mfa_version: 2 | |
| min_frames: 0 | |
| min_level_db: -100 | |
| noise_schedule: '' | |
| num_ckpt_keep: 3 | |
| num_heads: 2 | |
| num_mels: 80 | |
| num_sanity_val_steps: -1 | |
| num_spk: 400 | |
| num_test_samples: 0 | |
| num_valid_plots: 10 | |
| optimizer_adam_beta1: 0.9 | |
| optimizer_adam_beta2: 0.98 | |
| out_wav_norm: false | |
| pitch_extractor: parselmouth | |
| pre_align_args: | |
| allow_no_txt: false | |
| denoise: false | |
| nsample_per_mfa_group: 1000 | |
| sox_resample: false | |
| sox_to_wav: false | |
| trim_sil: false | |
| txt_processor: en | |
| use_tone: true | |
| pre_align_cls: egs.datasets.audio.pre_align.PreAlign | |
| print_nan_grads: false | |
| processed_data_dir: data/processed/LJSpeech | |
| profile_infer: false | |
| raw_data_dir: data/raw/LJSpeech-1.1 | |
| ref_level_db: 20 | |
| rename_tmux: true | |
| resume_from_checkpoint: 0 | |
| save_best: true | |
| save_codes: [] | |
| save_f0: false | |
| save_gt: true | |
| scheduler: rsqrt | |
| seed: 1234 | |
| sort_by_len: true | |
| task_cls: modules.FastDiff.task.FastDiff.FastDiffTask | |
| tb_log_interval: 100 | |
| test_ids: [] | |
| test_input_dir: '' | |
| test_mel_dir: '' | |
| test_num: 100 | |
| test_set_name: test | |
| train_set_name: train | |
| train_sets: '' | |
| upsample_ratios: | |
| - 8 | |
| - 8 | |
| - 4 | |
| use_pitch_embed: false | |
| use_spk_embed: false | |
| use_spk_id: false | |
| use_split_spk_id: false | |
| use_wav: true | |
| use_weight_norm: true | |
| use_word_input: false | |
| val_check_interval: 2000 | |
| valid_infer_interval: 10000 | |
| valid_monitor_key: val_loss | |
| valid_monitor_mode: min | |
| valid_set_name: valid | |
| vocoder_denoise_c: 0.0 | |
| warmup_updates: 8000 | |
| weight_decay: 0 | |
| win_length: null | |
| win_size: 1024 | |
| window: hann | |
| word_size: 30000 | |
| work_dir: checkpoints/FastDiff | |