hyperparams.yaml

# Generated 2023-06-20 from:
# /netscratch/sagar/thesis/speechbrain/recipes/RescueSpeech/Enhancement/fine-tuning/hparams/sepformer_16k.yaml
# yamllint disable
# ################################
# Model: SepFormer for source separation
# https://arxiv.org/abs/2010.13154
#
# Author:  Sangeet Sagar 2022
# Dataset : RescueSpeech
# ################################

# Basic parameters
# Seed needs to be set at top of yaml, before objects with parameters are made
seed: 8201
__set_seed: !apply:torch.manual_seed [8201]
experiment_name: sepformer-enhancement
output_folder: results/sepformer-enhancement/8201
train_log: results/sepformer-enhancement/8201/train_log.txt
save_folder: results/sepformer-enhancement/8201/save

# Dataset prep parameters
data_folder: dataset/audio_sythesis/Task_enhancement/ # !PLACEHOLDER
csv_dir: csv_files
train_csv: csv_files/train.csv
valid_csv: csv_files/dev.csv
test_csv: csv_files/test.csv
skip_prep: false
sample_rate: 16000
task: enhance

dereverberate: false
shuffle_train_data: true

# Pretrained models
pretrained_model_path: 
  /netscratch/sagar/thesis/speechbrain/recipes/RescueSpeech/pre-trained/sepformer_dns_16k                      # !PLACEHOLDER # sepformer_dns_16k model

# Basic parameters
use_tensorboard: false
tensorboard_logs: results/sepformer-enhancement/8201/logs/

# Experiment params
auto_mix_prec: true     # Set it to True for mixed precision
test_only: false
num_spks: 1
noprogressbar: false
save_audio: true        # Save estimated sources on disk
downsample: false
n_audio_to_save: 500

# Training parameters
N_epochs: 150
batch_size: 1
batch_size_test: 1
lr: 0.00015
clip_grad_norm: 5
loss_upper_lim: 999999  # this is the upper limit for an acceptable loss
# if True, the training sequences are cut to a specified length
limit_training_signal_len: false
# this is the length of sequences if we choose to limit
# the signal length of training sequences
training_signal_len: 32000
ckpt_interval_minutes: 60

# Parameters for data augmentation
use_wavedrop: false
use_speedperturb: true
use_rand_shift: false
min_shift: -8000
max_shift: 8000

speedperturb: !new:speechbrain.lobes.augment.TimeDomainSpecAugment
  perturb_prob: 1.0
  drop_freq_prob: 0.0
  drop_chunk_prob: 0.0
  sample_rate: 16000
  speeds: [95, 100, 105]

wavedrop: !new:speechbrain.lobes.augment.TimeDomainSpecAugment
  perturb_prob: 0.0
  drop_freq_prob: 1.0
  drop_chunk_prob: 1.0
  sample_rate: 16000

# loss thresholding -- this thresholds the training loss
threshold_byloss: true
threshold: -30

# Encoder parameters
N_encoder_out: 256
out_channels: 256
kernel_size: 16
kernel_stride: 8

# Dataloader options
dataloader_opts:
  batch_size: 1
  num_workers: 3

dataloader_opts_valid:
  batch_size: 1
  num_workers: 3

dataloader_opts_test:
  batch_size: 1
  num_workers: 3

# Specifying the network
Encoder: &id003 !new:speechbrain.lobes.models.dual_path.Encoder
  kernel_size: 16
  out_channels: 256

SBtfintra: &id001 !new:speechbrain.lobes.models.dual_path.SBTransformerBlock
  num_layers: 8
  d_model: 256
  nhead: 8
  d_ffn: 1024
  dropout: 0
  use_positional_encoding: true
  norm_before: true

SBtfinter: &id002 !new:speechbrain.lobes.models.dual_path.SBTransformerBlock
  num_layers: 8
  d_model: 256
  nhead: 8
  d_ffn: 1024
  dropout: 0
  use_positional_encoding: true
  norm_before: true

MaskNet: &id005 !new:speechbrain.lobes.models.dual_path.Dual_Path_Model

  num_spks: 1
  in_channels: 256
  out_channels: 256
  num_layers: 2
  K: 250
  intra_model: *id001
  inter_model: *id002
  norm: ln
  linear_layer_after_inter_intra: false
  skip_around_intra: true

Decoder: &id004 !new:speechbrain.lobes.models.dual_path.Decoder
  in_channels: 256
  out_channels: 1
  kernel_size: 16
  stride: 8
  bias: false

optimizer: !name:torch.optim.Adam
  lr: 0.00015
  weight_decay: 0

loss: !name:speechbrain.nnet.losses.get_si_snr_with_pitwrapper

lr_scheduler: &id007 !new:speechbrain.nnet.schedulers.ReduceLROnPlateau

  factor: 0.5
  patience: 2
  dont_halve_until_epoch: 85

epoch_counter: &id006 !new:speechbrain.utils.epoch_loop.EpochCounter
  limit: 150

modules:
  encoder: *id003
  decoder: *id004
  masknet: *id005
save_all_checkpoints: false
checkpointer: !new:speechbrain.utils.checkpoints.Checkpointer
  checkpoints_dir: results/sepformer-enhancement/8201/save
  recoverables:
    encoder: *id003
    decoder: *id004
    masknet: *id005
    counter: *id006
    lr_scheduler: *id007
train_logger: !new:speechbrain.utils.train_logger.FileTrainLogger
  save_file: results/sepformer-enhancement/8201/train_log.txt

## Uncomment if you wish to fine-tune a pre-trained model.
pretrainer: !new:speechbrain.utils.parameter_transfer.Pretrainer
  loadables:
    encoder: !ref <Encoder>
    masknet: !ref <MaskNet>
    decoder: !ref <Decoder>