---
tags:
- espnet
- audio
---


Repository: `ms180/mini_an4_integration_test`


## Training

- System: `ASRSystem`
- Recipe: `mini_an4/asr`
- Creator: `ms180`
- Created: `2026-01-14T00:36:01.951598`
- Git: `8509faad9811b58d5024f29fb9d68ffb026b5e73` (dirty)

## Pack

- Archive: `model_pack`
- Strategy: `espnet3`
- Exp dir: `exp/train_asr_rnn_data_aug_debug`


## Train config

<details><summary>expand</summary>

```
num_device: 1
num_nodes: 1
task: espnet3.systems.asr.task.ASRTask
recipe_dir: .
data_dir: ./data
exp_tag: train_asr_rnn_data_aug_debug
exp_dir: ./exp/train_asr_rnn_data_aug_debug
stats_dir: ./exp/stats
decode_dir: ./exp/train_asr_rnn_data_aug_debug/decode
dataset_dir: ./data/mini_an4
create_dataset:
  func: src.create_dataset.create_dataset
  dataset_dir: ./data/mini_an4
  archive_path: ./../../egs2/mini_an4/asr1/downloads.tar.gz
dataset:
  _target_: espnet3.components.data.data_organizer.DataOrganizer
  train:
  - name: train_nodev
    dataset:
      _target_: src.dataset.MiniAN4Dataset
      manifest_path: ./data/mini_an4/manifest/train_nodev.tsv
  valid:
  - name: train_dev
    dataset:
      _target_: src.dataset.MiniAN4Dataset
      manifest_path: ./data/mini_an4/manifest/train_dev.tsv
  preprocessor:
    _target_: espnet2.train.preprocessor.CommonPreprocessor
    _convert_: all
    fs: 16000
    train: true
    data_aug_effects:
    - - 0.1
      - contrast
      - enhancement_amount: 75.0
    - - 0.1
      - highpass
      - cutoff_freq: 5000
        Q: 0.707
    - - 0.1
      - equalization
      - center_freq: 1000
        gain: 0
        Q: 0.707
    - - 0.1
      - - - 0.3
          - speed_perturb
          - factor: 0.9
        - - 0.3
          - speed_perturb
          - factor: 1.1
        - - 0.3
          - speed_perturb
          - factor: 1.3
    data_aug_num:
    - 1
    - 4
    data_aug_prob: 1.0
    token_type: bpe
    token_list: ./data/bpe_30/tokens.txt
    bpemodel: ./data/bpe_30/bpe.model
parallel:
  env: local
  n_workers: 1
dataloader:
  collate_fn:
    _target_: espnet2.train.collate_fn.CommonCollateFn
    int_pad_value: -1
  train:
    multiple_iterator: false
    num_shards: 1
    iter_factory:
      _target_: espnet2.iterators.sequence_iter_factory.SequenceIterFactory
      shuffle: true
      collate_fn:
        _target_: espnet2.train.collate_fn.CommonCollateFn
        int_pad_value: -1
      num_workers: 0
      batches:
        type: sorted
        shape_files:
        - ./exp/stats/train/feats_shape
        batch_size: 2
        batch_bins: 200000
  valid:
    multiple_iterator: false
    num_shards: 1
    iter_factory:
      _target_: espnet2.iterators.sequence_iter_factory.SequenceIterFactory
      shuffle: false
      collate_fn:
        _target_: espnet2.train.collate_fn.CommonCollateFn
        int_pad_value: -1
      batches:
        type: sorted
        shape_files:
        - ./exp/stats/valid/feats_shape
        batch_size: 2
        batch_bins: 200000
optim:
  _target_: torch.optim.Adam
  lr: 0.001
  weight_decay: 0.0
scheduler:
  _target_: torch.optim.lr_scheduler.ReduceLROnPlateau
  mode: min
  factor: 0.5
  patience: 1
val_scheduler_criterion: valid/loss
best_model_criterion:
- - valid/acc
  - 1
  - max
trainer:
  accelerator: auto
  devices: 1
  num_nodes: 1
  accumulate_grad_batches: 1
  check_val_every_n_epoch: 1
  gradient_clip_val: 1.0
  log_every_n_steps: 1
  max_epochs: 1
  limit_train_batches: 1
  limit_val_batches: 1
  precision: 32
  logger:
  - _target_: lightning.pytorch.loggers.TensorBoardLogger
    save_dir: ./exp/train_asr_rnn_data_aug_debug/tensorboard
    name: tb_logger
  strategy: auto
tokenizer:
  vocab_size: 30
  character_coverage: 1.0
  model_type: bpe
  save_path: ./data/bpe_30
  text_builder:
    func: src.tokenizer.gather_training_text
    manifest_path: ./data/mini_an4/manifest/train_nodev.tsv
model:
  vocab_size: 30
  token_list: ./data/bpe_30/tokens.txt
  encoder: vgg_rnn
  encoder_conf:
    num_layers: 1
    hidden_size: 2
    output_size: 2
  decoder: rnn
  decoder_conf:
    hidden_size: 2
  normalize: utterance_mvn
  normalize_conf: {}
  model_conf:
    ctc_weight: 0.3
    lsm_weight: 0.1
    length_normalized_loss: false
  frontend: default
  frontend_conf:
    n_fft: 512
    win_length: 400
    hop_length: 160

```

</details>

### Citing ESPnet

```
@inproceedings{watanabe2018espnet,
  author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and
    Jiro Nishitoba and Yuya Unno and Nelson Yalta and Jahn Heymann and Matthew Wiesner
    and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai},
  title={{ESPnet}: End-to-End Speech Processing Toolkit},
  year={2018},
  booktitle={Proceedings of Interspeech},
  pages={2207--2211},
  doi={10.21437/Interspeech.2018-1456}
}
```