train_hf_wav2vec.yaml

# ################################
# Model: wav2vec2 + DNN + CTC
# Decoding AM: Greedy for validation, and Beam search for testing
# Augmentation: SpecAugment
# Authors: Sung-Lin Yeh 2021, Adel Moumen 2023
# ################################

# Seed needs to be set at top of yaml, before objects with parameters are made
seed: 1986
__set_seed: !apply:speechbrain.utils.seed_everything [!ref <seed>]
output_folder: !ref results/train_wav2vec2_char/<seed>
output_wer_folder: !ref <output_folder>/
save_folder: !ref <output_folder>/save
train_log: !ref <output_folder>/train_log.txt

# URL for the biggest Fairseq english wav2vec2 model.
wav2vec2_hub: facebook/wav2vec2-large-960h-lv60-self
wav2vec2_folder: !ref <save_folder>/wav2vec2_checkpoint

# Data files
data_folder: !PLACEHOLDER  # e.g., /path/to/LibriSpeech
# noise/ris dataset will automatically be downloaded
# data_folder_rirs: !ref <data_folder>
train_splits: ["train-clean-100", "train-clean-360", "train-other-500"]
dev_splits: ["dev-clean"]
test_splits: ["test-clean", "test-other"]
skip_prep: False
ckpt_interval_minutes: 25 # save checkpoint every N min
train_csv: !ref <output_folder>/train.csv
valid_csv: !ref <output_folder>/dev-clean.csv
test_csv:
   - !ref <output_folder>/test-clean.csv
   - !ref <output_folder>/test-other.csv

####################### Training Parameters ####################################

number_of_epochs: 1
lr: 0.9
lr_wav2vec: 0.0001
sorting: ascending
precision: fp32 # bf16, fp16 or fp32
sample_rate: 16000

# With data_parallel batch_size is split into N jobs
# With DDP batch_size is multiplied by N jobs
# Must be 3 per GPU to fit 32GB of VRAM
batch_size: 6
test_batch_size: 8

# Dataloader options
train_dataloader_opts:
   batch_size: !ref <batch_size>

valid_dataloader_opts:
   batch_size: !ref <batch_size>

test_dataloader_opts:
   batch_size: !ref <test_batch_size>

####################### Model Parameters #######################################
activation: !name:torch.nn.LeakyReLU
dnn_layers: 2
dnn_neurons: 1024
freeze_wav2vec: True

# Outputs
output_neurons: 29  # BPE size, index(blank/eos/bos) = 0
blank_index: 0

#
# Functions and classes
#

label_encoder: !new:speechbrain.dataio.encoder.CTCTextEncoder

epoch_counter: !new:speechbrain.utils.epoch_loop.EpochCounter
   limit: !ref <number_of_epochs>

enc: !new:speechbrain.lobes.models.VanillaNN.VanillaNN
   input_shape: [null, null, 1024]
   activation: !ref <activation>
   dnn_blocks: !ref <dnn_layers>
   dnn_neurons: !ref <dnn_neurons>

wav2vec2: !new:speechbrain.lobes.models.huggingface_transformers.wav2vec2.Wav2Vec2
   source: !ref <wav2vec2_hub>
   output_norm: True
   freeze: !ref <freeze_wav2vec>
   save_path: !ref <wav2vec2_folder>

#####
# Uncomment this block if you prefer to use a Fairseq pretrained model instead
# of a HuggingFace one. Here, we provide an URL that is obtained from the
# Fairseq github for the multilingual XLSR.
#
#wav2vec2_url: https://dl.fbaipublicfiles.com/fairseq/wav2vec/wav2vec_vox_960h_pl.pt
#wav2vec2: !new:speechbrain.lobes.models.fairseq_wav2vec.FairseqWav2Vec2
#    pretrained_path: !ref <wav2vec2_url>
#    output_norm: True
#    freeze: False
#    save_path: !ref <save_folder>/wav2vec2_checkpoint/model.pt

ctc_lin: !new:speechbrain.nnet.linear.Linear
   input_size: !ref <dnn_neurons>
   n_neurons: !ref <output_neurons>

log_softmax: !new:speechbrain.nnet.activations.Softmax
   apply_log: True

ctc_cost: !name:speechbrain.nnet.losses.ctc_loss
   blank_index: !ref <blank_index>

modules:
   wav2vec2: !ref <wav2vec2>
   enc: !ref <enc>
   ctc_lin: !ref <ctc_lin>

model: !new:torch.nn.ModuleList
   - [!ref <enc>, !ref <ctc_lin>]

model_opt_class: !name:torch.optim.Adadelta
   lr: !ref <lr>
   rho: 0.95
   eps: 1.e-8

wav2vec_opt_class: !name:torch.optim.Adam
   lr: !ref <lr_wav2vec>

lr_annealing_model: !new:speechbrain.nnet.schedulers.NewBobScheduler
   initial_value: !ref <lr>
   improvement_threshold: 0.0025
   annealing_factor: 0.8
   patient: 0

lr_annealing_wav2vec: !new:speechbrain.nnet.schedulers.NewBobScheduler
   initial_value: !ref <lr_wav2vec>
   improvement_threshold: 0.0025
   annealing_factor: 0.9
   patient: 0

############################## Augmentations ###################################

# Speed perturbation
speed_perturb: !new:speechbrain.augment.time_domain.SpeedPerturb
   orig_freq: !ref <sample_rate>
   speeds: [95, 100, 105]

# Frequency drop: randomly drops a number of frequency bands to zero.
drop_freq: !new:speechbrain.augment.time_domain.DropFreq
   drop_freq_low: 0
   drop_freq_high: 1
   drop_freq_count_low: 1
   drop_freq_count_high: 3
   drop_freq_width: 0.05

# Time drop: randomly drops a number of temporal chunks.
drop_chunk: !new:speechbrain.augment.time_domain.DropChunk
   drop_length_low: 1000
   drop_length_high: 2000
   drop_count_low: 1
   drop_count_high: 5

# Augmenter: Combines previously defined augmentations to perform data augmentation
wav_augment: !new:speechbrain.augment.augmenter.Augmenter
   concat_original: True
   min_augmentations: 4
   max_augmentations: 4
   augment_prob: 1.0
   augmentations: [
      !ref <speed_perturb>,
      !ref <drop_freq>,
      !ref <drop_chunk>]

############################## Decoding ########################################

# Decoding parameters
test_beam_search:
   beam_size: 143
   topk: 1
   blank_index: !ref <blank_index>
   space_token: ' ' # make sure this is the same as the one used in the tokenizer
   beam_prune_logp: -12.0
   token_prune_min_logp: -1.2
   prune_history: True
   alpha: 0.8
   beta: 1.2
   # can be downloaded from here https://www.openslr.org/11/ or trained with kenLM
   # It can either be a .bin or .arpa ; note: .arpa is much slower at loading
   # If you don't want to use an LM, comment it out or set it to null
   kenlm_model_path: null

############################## Logging and Pretrainer ##########################

checkpointer: !new:speechbrain.utils.checkpoints.Checkpointer
   checkpoints_dir: !ref <save_folder>
   recoverables:
      wav2vec2: !ref <wav2vec2>
      model: !ref <model>
      scheduler_model: !ref <lr_annealing_model>
      scheduler_wav2vec: !ref <lr_annealing_wav2vec>
      counter: !ref <epoch_counter>
      tokenizer: !ref <label_encoder>

train_logger: !new:speechbrain.utils.train_logger.FileTrainLogger
   save_file: !ref <train_log>

error_rate_computer: !name:speechbrain.utils.metric_stats.ErrorRateStats

cer_computer: !name:speechbrain.utils.metric_stats.ErrorRateStats
   split_tokens: True