| # This is the hyperparameter configuration file for FastSpeech2 v2. | |
| # the different of v2 and v1 is that v2 apply linformer technique. | |
| # Please make sure this is adjusted for the Baker dataset. If you want to | |
| # apply to the other dataset, you might need to carefully change some parameters. | |
| # This configuration performs 200k iters but a best checkpoint is around 150k iters. | |
| ########################################################### | |
| # FEATURE EXTRACTION SETTING # | |
| ########################################################### | |
| hop_size: 256 # Hop size. | |
| format: "npy" | |
| ########################################################### | |
| # NETWORK ARCHITECTURE SETTING # | |
| ########################################################### | |
| model_type: "fastspeech2" | |
| fastspeech2_params: | |
| dataset: baker | |
| n_speakers: 1 | |
| encoder_hidden_size: 256 | |
| encoder_num_hidden_layers: 3 | |
| encoder_num_attention_heads: 2 | |
| encoder_attention_head_size: 16 # in v1, = 384//2 | |
| encoder_intermediate_size: 1024 | |
| encoder_intermediate_kernel_size: 3 | |
| encoder_hidden_act: "mish" | |
| decoder_hidden_size: 256 | |
| decoder_num_hidden_layers: 3 | |
| decoder_num_attention_heads: 2 | |
| decoder_attention_head_size: 16 # in v1, = 384//2 | |
| decoder_intermediate_size: 1024 | |
| decoder_intermediate_kernel_size: 3 | |
| decoder_hidden_act: "mish" | |
| variant_prediction_num_conv_layers: 2 | |
| variant_predictor_filter: 256 | |
| variant_predictor_kernel_size: 3 | |
| variant_predictor_dropout_rate: 0.5 | |
| num_mels: 80 | |
| hidden_dropout_prob: 0.2 | |
| attention_probs_dropout_prob: 0.1 | |
| max_position_embeddings: 2048 | |
| initializer_range: 0.02 | |
| output_attentions: False | |
| output_hidden_states: False | |
| ########################################################### | |
| # DATA LOADER SETTING # | |
| ########################################################### | |
| batch_size: 16 # Batch size for each GPU with assuming that gradient_accumulation_steps == 1. | |
| remove_short_samples: true # Whether to remove samples the length of which are less than batch_max_steps. | |
| allow_cache: true # Whether to allow cache in dataset. If true, it requires cpu memory. | |
| mel_length_threshold: 32 # remove all targets has mel_length <= 32 | |
| is_shuffle: true # shuffle dataset after each epoch. | |
| ########################################################### | |
| # OPTIMIZER & SCHEDULER SETTING # | |
| ########################################################### | |
| optimizer_params: | |
| initial_learning_rate: 0.001 | |
| end_learning_rate: 0.00005 | |
| decay_steps: 150000 # < train_max_steps is recommend. | |
| warmup_proportion: 0.02 | |
| weight_decay: 0.001 | |
| gradient_accumulation_steps: 1 | |
| var_train_expr: null # trainable variable expr (eg. 'embeddings|encoder|decoder' ) | |
| # must separate by |. if var_train_expr is null then we | |
| # training all variable | |
| ########################################################### | |
| # INTERVAL SETTING # | |
| ########################################################### | |
| train_max_steps: 200000 # Number of training steps. | |
| save_interval_steps: 5000 # Interval steps to save checkpoint. | |
| eval_interval_steps: 500 # Interval steps to evaluate the network. | |
| log_interval_steps: 200 # Interval steps to record the training log. | |
| delay_f0_energy_steps: 3 # 2 steps use LR outputs only then 1 steps LR + F0 + Energy. | |
| ########################################################### | |
| # OTHER SETTING # | |
| ########################################################### | |
| num_save_intermediate_results: 1 # Number of batch to be saved as intermediate results. | |