Text-to-Speech
ONNX
zero-shot
multilingual
Approximetal's picture
Upload folder using huggingface_hub
2df45bb verified
{
"_name": null,
"common": {
"_name": null,
"no_progress_bar": false,
"log_interval": 100,
"log_format": "simple",
"log_file": null,
"aim_repo": null,
"aim_run_hash": null,
"tensorboard_logdir": "/checkpoint/mjhwang/experiments/230930-noiseaug_p2v-mls_multilingual_6lang/231005-noiseaug_p2v-mls_multilingual_6lang-alignfix.config_v2.langemb1.vuv_logit1.denoise.ngpu16",
"wandb_project": null,
"azureml_logging": false,
"seed": 1,
"cpu": false,
"tpu": false,
"bf16": false,
"fp16": false,
"memory_efficient_fp16": false,
"fp16_no_flatten_grads": false,
"fp16_init_scale": 128,
"fp16_scale_window": null,
"fp16_scale_tolerance": 0.0,
"on_cpu_convert_precision": false,
"min_loss_scale": 0.0001,
"threshold_loss_scale": null,
"amp": false,
"amp_batch_retries": 2,
"amp_init_scale": 128,
"amp_scale_window": null,
"user_dir": null,
"empty_cache_freq": 0,
"all_gather_list_size": 9999999,
"model_parallel_size": 1,
"quantization_config_path": null,
"profile": false,
"reset_logging": false,
"suppress_crashes": false,
"use_plasma_view": false,
"plasma_path": "/tmp/plasma",
"log_nvidia_smi": false,
"use_tutel_moe": false
},
"common_eval": {
"_name": null,
"path": null,
"post_process": null,
"quiet": false,
"model_overrides": "{}",
"results_path": null,
"is_moe": false,
"moe_generation": false
},
"distributed_training": {
"_name": null,
"distributed_world_size": 16,
"distributed_num_procs": 8,
"distributed_rank": 0,
"distributed_backend": "nccl",
"distributed_init_method": "tcp://learnfair0791:15129",
"distributed_port": 15129,
"device_id": 0,
"distributed_no_spawn": false,
"ddp_backend": "legacy_ddp",
"ddp_comm_hook": "none",
"bucket_cap_mb": 25,
"fix_batches_to_gpus": false,
"find_unused_parameters": true,
"gradient_as_bucket_view": false,
"fast_stat_sync": false,
"heartbeat_timeout": -1,
"broadcast_buffers": false,
"slowmo_momentum": null,
"slowmo_base_algorithm": "localsgd",
"localsgd_frequency": 3,
"nprocs_per_node": 8,
"pipeline_model_parallel": false,
"pipeline_balance": null,
"pipeline_devices": null,
"pipeline_chunks": 0,
"pipeline_encoder_balance": null,
"pipeline_encoder_devices": null,
"pipeline_decoder_balance": null,
"pipeline_decoder_devices": null,
"pipeline_checkpoint": "never",
"zero_sharding": "none",
"fp16": false,
"bf16": false,
"memory_efficient_fp16": false,
"tpu": false,
"no_reshard_after_forward": false,
"fp32_reduce_scatter": false,
"cpu_offload": false,
"use_sharded_state": false,
"not_fsdp_flatten_parameters": false,
"freeze_up_to_layer": null
},
"dataset": {
"_name": null,
"num_workers": 0,
"num_workers_valid": 0,
"skip_invalid_size_inputs_valid_test": true,
"max_tokens": 300000,
"batch_size": null,
"required_batch_size_multiple": 8,
"required_seq_len_multiple": 1,
"dataset_impl": null,
"data_buffer_size": 10,
"train_subset": "train_wenet_cmn_9_10,train_wenet_cmn_8_10,train_wenet_cmn_7_10,train_wenet_cmn_6_10,train_wenet_cmn_5_10,train_wenet_cmn_4_10,train_wenet_cmn_3_10,train_wenet_cmn_2_10,train_wenet_cmn_1_10,train_wenet_cmn_0_10,train_mls_en_9_10,train_mls_en_8_10,train_mls_en_7_10,train_mls_en_6_10,train_mls_en_5_10,train_mls_en_4_10,train_mls_en_3_10,train_mls_en_2_10,train_mls_en_1_10,train_mls_en_0_10,train_mls_deu,train_mls_fra,train_mls_spa,train_cv12_cmn,train_mls_ita,train_cv12_cmn_2,train_vl107_cmn",
"valid_subset": "dev_all",
"combine_valid_subsets": null,
"ignore_unused_valid_subsets": false,
"validate_interval": 1,
"validate_interval_updates": 5000,
"validate_after_updates": 0,
"fixed_validation_seed": null,
"disable_validation": false,
"max_tokens_valid": 300000,
"batch_size_valid": null,
"max_valid_steps": null,
"curriculum": 0,
"gen_subset": "test",
"num_shards": 1,
"shard_id": 0,
"grouped_shuffling": false,
"update_epoch_batch_itr": false,
"update_ordered_indices_seed": false
},
"optimization": {
"_name": null,
"max_epoch": 0,
"max_update": 500000,
"stop_time_hours": 0.0,
"clip_norm": 1.0,
"clip_norm_type": "l2",
"sentence_avg": false,
"update_freq": [
4
],
"lr": [
0.0001
],
"stop_min_lr": -1.0,
"use_bmuf": false,
"skip_remainder_batch": false
},
"checkpoint": {
"_name": null,
"save_dir": "/checkpoint/mjhwang/experiments/230930-noiseaug_p2v-mls_multilingual_6lang/231005-noiseaug_p2v-mls_multilingual_6lang-alignfix.config_v2.langemb1.vuv_logit1.denoise.ngpu16",
"restore_file": "checkpoint_last.pt",
"continue_once": null,
"finetune_from_model": null,
"ignore_suffix": false,
"reset_dataloader": true,
"reset_lr_scheduler": false,
"reset_meters": false,
"reset_optimizer": false,
"optimizer_overrides": "{}",
"save_interval": 1,
"save_interval_updates": 10000,
"keep_interval_updates": 1,
"keep_interval_updates_pattern": -1,
"keep_last_epochs": -1,
"keep_best_checkpoints": 10,
"no_save": false,
"no_epoch_checkpoints": true,
"no_last_checkpoints": false,
"no_best_checkpoints": false,
"no_save_optimizer_state": false,
"no_save_optimizer_state_on_training_finished": false,
"synchronize_checkpoints_before_copy": false,
"symlink_best_and_last_checkpoints": false,
"best_checkpoint_metric": "mse_loss",
"maximize_best_checkpoint_metric": false,
"patience": 20,
"checkpoint_suffix": "",
"checkpoint_shard_count": 1,
"load_checkpoint_on_all_dp_ranks": false,
"write_checkpoints_asynchronously": false,
"s3_upload_path": null,
"replication_count": 1,
"model_parallel_size": 1
},
"bmuf": {
"_name": null,
"block_lr": 1.0,
"block_momentum": 0.875,
"global_sync_iter": 50,
"warmup_iterations": 500,
"use_nbm": false,
"average_sync": false,
"distributed_world_size": 16
},
"generation": {
"_name": null,
"beam": 5,
"beam_mt": 0,
"nbest": 1,
"max_len_a": 0.0,
"max_len_b": 200,
"max_len_a_mt": 0.0,
"max_len_b_mt": 200,
"min_len": 1,
"match_source_len": false,
"unnormalized": false,
"no_early_stop": false,
"no_beamable_mm": false,
"lenpen": 1.0,
"lenpen_mt": 1.0,
"unkpen": 0.0,
"blankpen": 0.0,
"replace_unk": null,
"sacrebleu": false,
"score_reference": false,
"prefix_size": 0,
"no_repeat_ngram_size": 0,
"sampling": false,
"sampling_topk": -1,
"sampling_topp": -1.0,
"constraints": null,
"temperature": 1.0,
"diverse_beam_groups": -1,
"diverse_beam_strength": 0.5,
"diversity_rate": -1.0,
"print_alignment": null,
"print_step": false,
"lm_path": null,
"lm_weight": 0.0,
"iter_decode_eos_penalty": 0.0,
"iter_decode_max_iter": 10,
"iter_decode_force_max_iter": false,
"iter_decode_with_beam": 1,
"iter_decode_with_external_reranker": false,
"retain_iter_history": false,
"retain_dropout": false,
"retain_dropout_modules": null,
"decoding_format": null,
"no_seed_provided": false,
"eos_token": null
},
"eval_lm": {
"_name": null,
"output_word_probs": false,
"output_word_stats": false,
"context_window": 0,
"softmax_batch": 9223372036854775807,
"stats_path": null,
"max_valid_steps": null
},
"interactive": {
"_name": null,
"buffer_size": 0,
"input": "-"
},
"model": {
"no_progress_bar": false,
"log_interval": 100,
"log_format": "simple",
"log_file": null,
"aim_repo": null,
"aim_run_hash": null,
"tensorboard_logdir": "/checkpoint/mjhwang/experiments/230930-noiseaug_p2v-mls_multilingual_6lang/231005-noiseaug_p2v-mls_multilingual_6lang-alignfix.config_v2.langemb1.vuv_logit1.denoise.ngpu16",
"wandb_project": null,
"azureml_logging": false,
"seed": 1,
"cpu": false,
"tpu": false,
"bf16": false,
"fp16": false,
"memory_efficient_fp16": false,
"fp16_no_flatten_grads": false,
"fp16_init_scale": 128,
"fp16_scale_window": null,
"fp16_scale_tolerance": 0.0,
"on_cpu_convert_precision": false,
"min_loss_scale": 0.0001,
"threshold_loss_scale": null,
"amp": false,
"amp_batch_retries": 2,
"amp_init_scale": 128,
"amp_scale_window": null,
"user_dir": null,
"empty_cache_freq": 0,
"all_gather_list_size": 9999999,
"model_parallel_size": 1,
"quantization_config_path": null,
"profile": false,
"reset_logging": false,
"suppress_crashes": false,
"use_plasma_view": false,
"plasma_path": "/tmp/plasma",
"log_nvidia_smi": false,
"use_tutel_moe": false,
"tokenizer": null,
"bpe": null,
"optimizer": "adam",
"lr_scheduler": "fixed",
"simul_type": null,
"criterion": "nar_prosody2vec",
"scoring": "bleu",
"task": "prosody2vec",
"num_workers": 0,
"num_workers_valid": 0,
"skip_invalid_size_inputs_valid_test": true,
"max_tokens": 300000,
"batch_size": null,
"required_batch_size_multiple": 8,
"required_seq_len_multiple": 1,
"dataset_impl": null,
"data_buffer_size": 10,
"train_subset": "train_wenet_cmn_9_10,train_wenet_cmn_8_10,train_wenet_cmn_7_10,train_wenet_cmn_6_10,train_wenet_cmn_5_10,train_wenet_cmn_4_10,train_wenet_cmn_3_10,train_wenet_cmn_2_10,train_wenet_cmn_1_10,train_wenet_cmn_0_10,train_mls_en_9_10,train_mls_en_8_10,train_mls_en_7_10,train_mls_en_6_10,train_mls_en_5_10,train_mls_en_4_10,train_mls_en_3_10,train_mls_en_2_10,train_mls_en_1_10,train_mls_en_0_10,train_mls_deu,train_mls_fra,train_mls_spa,train_cv12_cmn,train_mls_ita,train_cv12_cmn_2,train_vl107_cmn",
"valid_subset": "dev_all",
"combine_valid_subsets": null,
"ignore_unused_valid_subsets": false,
"validate_interval": 1,
"validate_interval_updates": 5000,
"validate_after_updates": 0,
"fixed_validation_seed": null,
"disable_validation": false,
"max_tokens_valid": "300000",
"batch_size_valid": null,
"max_valid_steps": null,
"curriculum": 0,
"gen_subset": "test",
"num_shards": 1,
"shard_id": 0,
"grouped_shuffling": false,
"update_epoch_batch_itr": false,
"update_ordered_indices_seed": false,
"distributed_world_size": 16,
"distributed_num_procs": 8,
"distributed_rank": 0,
"distributed_backend": "nccl",
"distributed_init_method": null,
"distributed_port": 15129,
"device_id": 0,
"distributed_no_spawn": false,
"ddp_backend": "legacy_ddp",
"ddp_comm_hook": "none",
"bucket_cap_mb": 25,
"fix_batches_to_gpus": false,
"find_unused_parameters": true,
"gradient_as_bucket_view": false,
"fast_stat_sync": false,
"heartbeat_timeout": -1,
"broadcast_buffers": false,
"slowmo_momentum": null,
"slowmo_base_algorithm": "localsgd",
"localsgd_frequency": 3,
"nprocs_per_node": 8,
"pipeline_model_parallel": false,
"pipeline_balance": null,
"pipeline_devices": null,
"pipeline_chunks": 0,
"pipeline_encoder_balance": null,
"pipeline_encoder_devices": null,
"pipeline_decoder_balance": null,
"pipeline_decoder_devices": null,
"pipeline_checkpoint": "never",
"zero_sharding": "none",
"no_reshard_after_forward": false,
"fp32_reduce_scatter": false,
"cpu_offload": false,
"use_sharded_state": false,
"not_fsdp_flatten_parameters": false,
"freeze_up_to_layer": null,
"arch": "nar_p2v",
"max_epoch": 0,
"max_update": 500000,
"stop_time_hours": 0,
"clip_norm": 1.0,
"clip_norm_type": "l2",
"sentence_avg": false,
"update_freq": [
4
],
"lr": [
0.0001
],
"stop_min_lr": -1.0,
"use_bmuf": false,
"skip_remainder_batch": false,
"save_dir": "/checkpoint/mjhwang/experiments/230930-noiseaug_p2v-mls_multilingual_6lang/231005-noiseaug_p2v-mls_multilingual_6lang-alignfix.config_v2.langemb1.vuv_logit1.denoise.ngpu16",
"restore_file": "checkpoint_last.pt",
"continue_once": null,
"finetune_from_model": null,
"ignore_suffix": false,
"reset_dataloader": true,
"reset_lr_scheduler": false,
"reset_meters": false,
"reset_optimizer": false,
"optimizer_overrides": "{}",
"save_interval": 1,
"save_interval_updates": 10000,
"keep_interval_updates": 1,
"keep_interval_updates_pattern": -1,
"keep_last_epochs": -1,
"keep_best_checkpoints": 10,
"no_save": false,
"no_epoch_checkpoints": true,
"no_last_checkpoints": false,
"no_best_checkpoints": false,
"no_save_optimizer_state": false,
"no_save_optimizer_state_on_training_finished": false,
"synchronize_checkpoints_before_copy": false,
"symlink_best_and_last_checkpoints": false,
"best_checkpoint_metric": "mse_loss",
"maximize_best_checkpoint_metric": false,
"patience": 20,
"checkpoint_suffix": "",
"checkpoint_shard_count": 1,
"load_checkpoint_on_all_dp_ranks": false,
"write_checkpoints_asynchronously": false,
"s3_upload_path": null,
"replication_count": 1,
"store_ema": false,
"ema_decay": 0.9999,
"ema_start_update": 0,
"ema_seed_model": null,
"ema_update_freq": 1,
"ema_fp32": false,
"load_prosody_encoder_from": null,
"freeze_prosody_encoder": false,
"unit_encoder_arch": "daft_exprt_encoder",
"prosody_encoder_arch": "ecapa_tdnn2",
"decoder_arch": "daft_exprt_decoder",
"data": "/large_experiments/seamless/ust/mjhwang/data/denoise_prosody2vec/mls_multilingual_6lang_xlsr_10k_noiseaug",
"config_yaml": "config_v2.yaml",
"max_source_positions": 300000,
"max_target_positions": 300000,
"n_frames_per_step": 1,
"eos_prob_threshold": 0.5,
"eval_inference": true,
"eval_tb_nsample": 8,
"eval_bleu": false,
"vocoder": "griffin_lim",
"spec_bwd_max_iter": 8,
"jit_data_offloading": true,
"jit_data_root": "/scratch/slurm_tmpdir/${SLURM_JOB_ID}",
"adam_betas": "(0.9, 0.98)",
"adam_eps": 1e-08,
"weight_decay": 0.0,
"use_old_adam": false,
"fp16_adam_stats": false,
"block_wise": false,
"force_anneal": null,
"lr_shrink": 0.1,
"warmup_updates": 1000,
"ctc_weight": 0.0,
"forward_sum_weight": 1.0,
"bin_loss_start_ratio": 0.1,
"bin_loss_warmup_steps": 6000,
"film_regul_weight": 0.001,
"pros_consist_weight": 0.0,
"denoise_target": true,
"snr_threshold": 2000000000000000.0,
"pad": 1,
"eos": 2,
"unk": 3,
"use_spkr_emb": 0,
"use_lang_emb": 1,
"prosody_embed_dim": 512,
"use_ucmvn": 0,
"use_spec_augment": 1,
"use_prosody_layernorm": 1,
"var_pred_hidden_dim": 512,
"var_pred_kernel_size": 5,
"var_pred_n_bins": -1,
"add_variance_parallel": 1,
"use_film_decoder": 1,
"predict_var_vuv": 1,
"predict_vuv_logit": 1,
"predict_frm_f0_vuv": 0,
"no_seed_provided": false,
"speaker_embed_dim": 192,
"use_utterance_speaker_embed": false,
"lang_embed_dim": 64,
"_name": "nar_p2v",
"lang_to_id": {
"cmn": 0,
"deu": 1,
"eng": 2,
"fra": 3,
"ita": 4,
"spa": 5
},
"pitch_min": 0.0,
"pitch_max": 6.858574643755327,
"energy_min": 0.0,
"energy_max": 6.360039234161377,
"speaker_emb_path": null,
"input_feat_per_channel": 80,
"input_channels": 1,
"speaker_to_id": null,
"dropout": 0.2,
"fft_hidden_dim": 1024,
"fft_kernel_size": 9,
"attention_dropout": 0.0,
"encoder_layers": 4,
"encoder_embed_dim": 256,
"encoder_attention_heads": 2,
"output_frame_dim": 80,
"prosody_channels": [
512,
512,
512,
512,
1536
],
"prosody_kernel_sizes": [
5,
3,
3,
3,
1
],
"prosody_dilations": [
1,
2,
3,
4,
1
],
"prosody_attention_channels": 128,
"prosody_res2net_scale": 8,
"prosody_se_channels": 128,
"prosody_global_context": true,
"prosody_groups": [
1,
1,
1,
1,
1
],
"decoder_layers": 4,
"decoder_embed_dim": 256,
"decoder_attention_heads": 2,
"var_pred_dropout": 0.5,
"add_postnet": true,
"postnet_dropout": 0.5,
"postnet_layers": 5,
"postnet_conv_dim": 512,
"postnet_conv_kernel_size": 5,
"upsampling": "gaussian"
},
"task": {
"no_progress_bar": false,
"log_interval": 100,
"log_format": "simple",
"log_file": null,
"aim_repo": null,
"aim_run_hash": null,
"tensorboard_logdir": "/checkpoint/mjhwang/experiments/230930-noiseaug_p2v-mls_multilingual_6lang/231005-noiseaug_p2v-mls_multilingual_6lang-alignfix.config_v2.langemb1.vuv_logit1.denoise.ngpu16",
"wandb_project": null,
"azureml_logging": false,
"seed": 1,
"cpu": false,
"tpu": false,
"bf16": false,
"fp16": false,
"memory_efficient_fp16": false,
"fp16_no_flatten_grads": false,
"fp16_init_scale": 128,
"fp16_scale_window": null,
"fp16_scale_tolerance": 0.0,
"on_cpu_convert_precision": false,
"min_loss_scale": 0.0001,
"threshold_loss_scale": null,
"amp": false,
"amp_batch_retries": 2,
"amp_init_scale": 128,
"amp_scale_window": null,
"user_dir": null,
"empty_cache_freq": 0,
"all_gather_list_size": 9999999,
"model_parallel_size": 1,
"quantization_config_path": null,
"profile": false,
"reset_logging": false,
"suppress_crashes": false,
"use_plasma_view": false,
"plasma_path": "/tmp/plasma",
"log_nvidia_smi": false,
"use_tutel_moe": false,
"tokenizer": null,
"bpe": null,
"optimizer": "adam",
"lr_scheduler": "fixed",
"simul_type": null,
"criterion": "nar_prosody2vec",
"scoring": "bleu",
"task": "prosody2vec",
"num_workers": 0,
"num_workers_valid": 0,
"skip_invalid_size_inputs_valid_test": true,
"max_tokens": 300000,
"batch_size": null,
"required_batch_size_multiple": 8,
"required_seq_len_multiple": 1,
"dataset_impl": null,
"data_buffer_size": 10,
"train_subset": "train_wenet_cmn_9_10,train_wenet_cmn_8_10,train_wenet_cmn_7_10,train_wenet_cmn_6_10,train_wenet_cmn_5_10,train_wenet_cmn_4_10,train_wenet_cmn_3_10,train_wenet_cmn_2_10,train_wenet_cmn_1_10,train_wenet_cmn_0_10,train_mls_en_9_10,train_mls_en_8_10,train_mls_en_7_10,train_mls_en_6_10,train_mls_en_5_10,train_mls_en_4_10,train_mls_en_3_10,train_mls_en_2_10,train_mls_en_1_10,train_mls_en_0_10,train_mls_deu,train_mls_fra,train_mls_spa,train_cv12_cmn,train_mls_ita,train_cv12_cmn_2,train_vl107_cmn",
"valid_subset": "dev_all",
"combine_valid_subsets": null,
"ignore_unused_valid_subsets": false,
"validate_interval": 1,
"validate_interval_updates": 5000,
"validate_after_updates": 0,
"fixed_validation_seed": null,
"disable_validation": false,
"max_tokens_valid": "300000",
"batch_size_valid": null,
"max_valid_steps": null,
"curriculum": 0,
"gen_subset": "test",
"num_shards": 1,
"shard_id": 0,
"grouped_shuffling": false,
"update_epoch_batch_itr": false,
"update_ordered_indices_seed": false,
"distributed_world_size": 16,
"distributed_num_procs": 8,
"distributed_rank": 0,
"distributed_backend": "nccl",
"distributed_init_method": null,
"distributed_port": 15129,
"device_id": 0,
"distributed_no_spawn": false,
"ddp_backend": "legacy_ddp",
"ddp_comm_hook": "none",
"bucket_cap_mb": 25,
"fix_batches_to_gpus": false,
"find_unused_parameters": true,
"gradient_as_bucket_view": false,
"fast_stat_sync": false,
"heartbeat_timeout": -1,
"broadcast_buffers": false,
"slowmo_momentum": null,
"slowmo_base_algorithm": "localsgd",
"localsgd_frequency": 3,
"nprocs_per_node": 8,
"pipeline_model_parallel": false,
"pipeline_balance": null,
"pipeline_devices": null,
"pipeline_chunks": 0,
"pipeline_encoder_balance": null,
"pipeline_encoder_devices": null,
"pipeline_decoder_balance": null,
"pipeline_decoder_devices": null,
"pipeline_checkpoint": "never",
"zero_sharding": "none",
"no_reshard_after_forward": false,
"fp32_reduce_scatter": false,
"cpu_offload": false,
"use_sharded_state": false,
"not_fsdp_flatten_parameters": false,
"freeze_up_to_layer": null,
"arch": "nar_p2v",
"max_epoch": 0,
"max_update": 500000,
"stop_time_hours": 0,
"clip_norm": 1.0,
"clip_norm_type": "l2",
"sentence_avg": false,
"update_freq": [
4
],
"lr": [
0.0001
],
"stop_min_lr": -1.0,
"use_bmuf": false,
"skip_remainder_batch": false,
"save_dir": "/checkpoint/mjhwang/experiments/230930-noiseaug_p2v-mls_multilingual_6lang/231005-noiseaug_p2v-mls_multilingual_6lang-alignfix.config_v2.langemb1.vuv_logit1.denoise.ngpu16",
"restore_file": "checkpoint_last.pt",
"continue_once": null,
"finetune_from_model": null,
"ignore_suffix": false,
"reset_dataloader": true,
"reset_lr_scheduler": false,
"reset_meters": false,
"reset_optimizer": false,
"optimizer_overrides": "{}",
"save_interval": 1,
"save_interval_updates": 10000,
"keep_interval_updates": 1,
"keep_interval_updates_pattern": -1,
"keep_last_epochs": -1,
"keep_best_checkpoints": 10,
"no_save": false,
"no_epoch_checkpoints": true,
"no_last_checkpoints": false,
"no_best_checkpoints": false,
"no_save_optimizer_state": false,
"no_save_optimizer_state_on_training_finished": false,
"synchronize_checkpoints_before_copy": false,
"symlink_best_and_last_checkpoints": false,
"best_checkpoint_metric": "mse_loss",
"maximize_best_checkpoint_metric": false,
"patience": 20,
"checkpoint_suffix": "",
"checkpoint_shard_count": 1,
"load_checkpoint_on_all_dp_ranks": false,
"write_checkpoints_asynchronously": false,
"s3_upload_path": null,
"replication_count": 1,
"store_ema": false,
"ema_decay": 0.9999,
"ema_start_update": 0,
"ema_seed_model": null,
"ema_update_freq": 1,
"ema_fp32": false,
"load_prosody_encoder_from": null,
"freeze_prosody_encoder": false,
"unit_encoder_arch": "daft_exprt_encoder",
"prosody_encoder_arch": "ecapa_tdnn2",
"decoder_arch": "daft_exprt_decoder",
"data": "/large_experiments/seamless/ust/mjhwang/data/denoise_prosody2vec/mls_multilingual_6lang_xlsr_10k_noiseaug",
"config_yaml": "config_v2.yaml",
"max_source_positions": 300000,
"max_target_positions": 300000,
"n_frames_per_step": 1,
"eos_prob_threshold": 0.5,
"eval_inference": true,
"eval_tb_nsample": 8,
"eval_bleu": false,
"vocoder": "griffin_lim",
"spec_bwd_max_iter": 8,
"jit_data_offloading": true,
"jit_data_root": "/scratch/slurm_tmpdir/${SLURM_JOB_ID}",
"adam_betas": "(0.9, 0.98)",
"adam_eps": 1e-08,
"weight_decay": 0.0,
"use_old_adam": false,
"fp16_adam_stats": false,
"block_wise": false,
"force_anneal": null,
"lr_shrink": 0.1,
"warmup_updates": 1000,
"ctc_weight": 0.0,
"forward_sum_weight": 1.0,
"bin_loss_start_ratio": 0.1,
"bin_loss_warmup_steps": 6000,
"film_regul_weight": 0.001,
"pros_consist_weight": 0.0,
"denoise_target": true,
"snr_threshold": 2000000000000000.0,
"pad": 1,
"eos": 2,
"unk": 3,
"use_spkr_emb": 0,
"use_lang_emb": 1,
"prosody_embed_dim": 512,
"use_ucmvn": 0,
"use_spec_augment": 1,
"use_prosody_layernorm": 1,
"var_pred_hidden_dim": 512,
"var_pred_kernel_size": 5,
"var_pred_n_bins": -1,
"add_variance_parallel": 1,
"use_film_decoder": 1,
"predict_var_vuv": 1,
"predict_vuv_logit": 1,
"predict_frm_f0_vuv": 0,
"no_seed_provided": false,
"speaker_embed_dim": 192,
"use_utterance_speaker_embed": false,
"lang_embed_dim": 64,
"_name": "prosody2vec"
},
"criterion": {
"_name": "nar_prosody2vec",
"ctc_weight": 0.0,
"forward_sum_weight": 1.0,
"bin_loss_start_ratio": 0.1,
"bin_loss_warmup_steps": 6000,
"film_regul_weight": 0.001,
"pros_consist_weight": 0.0,
"denoise_target": true,
"snr_threshold": 2000000000000000.0
},
"optimizer": {
"_name": "adam",
"adam_betas": "(0.9, 0.98)",
"adam_eps": 1e-08,
"weight_decay": 0.0,
"use_old_adam": false,
"fp16_adam_stats": false,
"tpu": false,
"lr": [
0.0001
],
"block_wise": false
},
"lr_scheduler": {
"_name": "fixed",
"force_anneal": null,
"lr_shrink": 0.1,
"warmup_updates": 1000,
"lr": [
0.0001
]
},
"scoring": {
"_name": "bleu",
"pad": 1,
"eos": 2,
"unk": 3
},
"bpe": null,
"tokenizer": null,
"ema": {
"_name": null,
"store_ema": false,
"ema_decay": 0.9999,
"ema_start_update": 0,
"ema_seed_model": null,
"ema_update_freq": 1,
"ema_fp32": false
},
"simul_type": null
}