metadata
language:
- 'no'
license: apache-2.0
base_model: NbAiLab/nb-whisper-small-v0.8-vad3
tags:
- audio
- asr
- automatic-speech-recognition
- hf-asr-leaderboard
model-index:
- name: nb-whisper-small-v0.8-vad3-verbatim
results: []
nb-whisper-small-v0.8-vad3-verbatim
This model is a fine-tuned version of NbAiLab/nb-whisper-small-v0.8-vad3 on the NbAiLab/NPSC dataset. It achieves the following results on the evaluation set:
- step: 249
- validation_loss: 0.5783
- train_loss: 0.4395
- validation_wer: 9.3843
- validation_cer: 3.1287
- validation_exact_wer: 9.5329
- validation_exact_cer: 3.1513
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: 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 | 1.4467 | 1.3978 | 19.2607 | 12.0780 | 34.1889 | 15.2244 |
| 40 | 0.5979 | 0.5311 | 10.8048 | 3.6534 | 10.9488 | 3.6881 |
| 80 | 0.5743 | 0.4963 | 9.9323 | 3.3896 | 10.0745 | 3.4130 |
| 120 | 0.5719 | 0.4591 | 9.8317 | 3.3754 | 9.9447 | 3.3959 |
| 160 | 0.5738 | 0.4571 | 9.5129 | 3.2325 | 9.6852 | 3.2569 |
| 200 | 0.5772 | 0.4494 | 9.4178 | 3.1392 | 9.5668 | 3.1618 |
| 240 | 0.5788 | 0.4449 | 9.3954 | 3.1192 | 9.5386 | 3.1418 |
| 249 | 0.5783 | 0.4395 | 9.3843 | 3.1287 | 9.5329 | 3.1513 |
Framework versions
- Transformers 4.34.1
- Datasets 2.16.1
- Tokenizers 0.14.1