| language: | |
| - en | |
| tags: | |
| - esc | |
| datasets: | |
| - chime4 | |
| To reproduce this run, execute: | |
| ```python | |
| #!/usr/bin/env bash | |
| python run_flax_speech_recognition_seq2seq.py \ | |
| --dataset_name="esc-benchmark/esc-datasets" \ | |
| --model_name_or_path="esc-benchmark/wav2vec2-aed-pretrained" \ | |
| --dataset_config_name="chime4" \ | |
| --output_dir="./" \ | |
| --wandb_name="wav2vec2-aed-chime4" \ | |
| --wandb_project="wav2vec2-aed" \ | |
| --per_device_train_batch_size="8" \ | |
| --per_device_eval_batch_size="4" \ | |
| --logging_steps="25" \ | |
| --max_steps="50001" \ | |
| --eval_steps="10000" \ | |
| --save_steps="10000" \ | |
| --generation_max_length="40" \ | |
| --generation_num_beams="1" \ | |
| --final_generation_max_length="250" \ | |
| --final_generation_num_beams="5" \ | |
| --generation_length_penalty="0.6" \ | |
| --learning_rate="1e-4" \ | |
| --warmup_steps="500" \ | |
| --hidden_dropout="0.2" \ | |
| --activation_dropout="0.2" \ | |
| --feat_proj_dropout="0.2" \ | |
| --overwrite_output_dir \ | |
| --gradient_checkpointing \ | |
| --freeze_feature_encoder \ | |
| --predict_with_generate \ | |
| --do_eval \ | |
| --do_train \ | |
| --do_predict \ | |
| --push_to_hub \ | |
| --use_auth_token | |
| ``` | |