metadata
language:
- 'no'
license: apache-2.0
tags:
- audio
- asr
- automatic-speech-recognition
- hf-asr-leaderboard
model-index:
- name: scream_small_beta
results: []
scream_small_beta
This model is a fine-tuned version of openai/whisper-small on the NbAiLab/ncc_speech dataset.
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- lr_scheduler_type: linear
- per_device_train_batch_size: 32
- total_train_batch_size_per_node: 128
- total_train_batch_size: 1024
- total_optimization_steps: 25,000
- starting_optimization_step: None
- finishing_optimization_step: 25,000
- num_train_dataset_workers: 32
- num_hosts: 8
- total_num_training_examples: 25,600,000
- steps_per_epoch: To be computed after first epoch
- num_beams: None
- dropout: True
- bpe_dropout_probability: 0.1
Training results
| step | validation_fleurs_loss | train_loss | validation_fleurs_wer | validation_fleurs_cer | validation_fleurs_exact_wer | validation_fleurs_exact_cer | validation_stortinget_loss | validation_stortinget_wer | validation_stortinget_cer | validation_stortinget_exact_wer | validation_stortinget_exact_cer | validation_nrk_tv_loss | validation_nrk_tv_wer | validation_nrk_tv_cer | validation_nrk_tv_exact_wer | validation_nrk_tv_exact_cer |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 1.2013 | 2.7117 | 32.3914 | 9.8343 | 35.7228 | 10.9398 | 1.4988 | 44.0673 | 22.9444 | 48.2612 | 24.2595 | 1.8165 | 79.9390 | 54.6020 | 89.7612 | 56.8482 |
| 1000 | 0.5796 | 1.0147 | 16.1214 | 5.2624 | 19.9821 | 6.2962 | 0.4822 | 22.0502 | 13.3652 | 25.7586 | 14.0827 | 1.0170 | 51.9187 | 37.4011 | 59.7853 | 39.0187 |
Framework versions
- Transformers 4.31.0.dev0
- Datasets 2.13.0
- Tokenizers 0.13.3