File size: 1,082 Bytes
d2f8541 f57d2bd d2f8541 f57d2bd d2f8541 f57d2bd d2f8541 f57d2bd d2f8541 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 |
---
language:
- en
tags:
- esb
datasets:
- esb/datasets
- kensho/spgispeech
---
To reproduce this run, first call `get_ctc_tokenizer.py` to train the CTC tokenizer and then execute the following command to train the CTC system:
```python
#!/usr/bin/env bash
python run_flax_speech_recognition_ctc.py \
--model_name_or_path="esb/wav2vec2-ctc-pretrained" \
--tokenizer_name="wav2vec2-ctc-spgispeech-tokenizer" \
--dataset_name="esb/datasets" \
--dataset_config_name="spgispeech" \
--output_dir="./" \
--wandb_project="wav2vec2-ctc" \
--wandb_name="wav2vec2-ctc-spgispeech" \
--max_steps="50000" \
--save_steps="10000" \
--eval_steps="10000" \
--learning_rate="3e-4" \
--logging_steps="25" \
--warmup_steps="5000" \
--preprocessing_num_workers="1" \
--do_train \
--do_eval \
--do_predict \
--overwrite_output_dir \
--gradient_checkpointing \
--freeze_feature_encoder \
--push_to_hub \
--use_auth_token
```
|