metadata
language:
- 'no'
license: apache-2.0
base_model: NbAiLab/nb-whisper-base-v0.8-vad3
tags:
- audio
- asr
- automatic-speech-recognition
- hf-asr-leaderboard
model-index:
- name: nb-whisper-base-v0.8-vad3-verbatim
results: []
nb-whisper-base-v0.8-vad3-verbatim
This model is a fine-tuned version of NbAiLab/nb-whisper-base-v0.8-vad3 on the NbAiLab/NPSC dataset. It achieves the following results on the evaluation set:
- step: 249
- validation_loss: 0.5419
- train_loss: 0.4718
- validation_wer: 11.3249
- validation_cer: 3.9000
- validation_exact_wer: 11.5693
- validation_exact_cer: 3.9375
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: 0.0001
- 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: 250
- starting_optimization_step: None
- finishing_optimization_step: 250
- num_train_dataset_workers: 32
- num_hosts: 8
- total_num_training_examples: 256,000
- steps_per_epoch: 45
- num_beams: None
- weight_decay: 0.01
- adam_beta1: 0.9
- adam_beta2: 0.98
- adam_epsilon: 1e-06
- dropout: True
- bpe_dropout_probability: 0.2
- activation_dropout_probability: 0.1
Training results
| step | validation_loss | train_loss | validation_wer | validation_cer | validation_exact_wer | validation_exact_cer |
|---|---|---|---|---|---|---|
| 0 | 1.2371 | 1.2082 | 21.0503 | 11.5619 | 35.6160 | 14.7314 |
| 40 | 0.5799 | 0.5877 | 13.1592 | 4.4751 | 13.4138 | 4.5352 |
| 80 | 0.5521 | 0.5398 | 11.8506 | 4.0542 | 12.0939 | 4.0935 |
| 120 | 0.5469 | 0.4995 | 11.6884 | 3.9923 | 11.9641 | 4.0345 |
| 160 | 0.5441 | 0.4875 | 11.3864 | 3.9305 | 11.6257 | 3.9651 |
| 200 | 0.5422 | 0.4770 | 11.3808 | 3.9209 | 11.6370 | 3.9631 |
| 240 | 0.5417 | 0.4789 | 11.2913 | 3.8886 | 11.5241 | 3.9251 |
| 249 | 0.5419 | 0.4718 | 11.3249 | 3.9000 | 11.5693 | 3.9375 |
Framework versions
- Transformers 4.34.1
- Datasets 2.16.1
- Tokenizers 0.14.1