metadata

language:
  - 'no'
license: apache-2.0
base_model: NbAiLab/nb-whisper-large-v3-RC4
tags:
  - audio
  - asr
  - automatic-speech-recognition
  - hf-asr-leaderboard
model-index:
  - name: nb-whisper-large-v0.7
    results: []

nb-whisper-large-v0.7

This model is a fine-tuned version of NbAiLab/nb-whisper-large-v3-RC4 on the NbAiLab/ncc_speech_styling_v2 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: 7e-05
lr_scheduler_type: linear
per_device_train_batch_size: 8
total_train_batch_size_per_node: 32
total_train_batch_size: 1024
total_optimization_steps: 50,000
starting_optimization_step: None
finishing_optimization_step: 50,000
num_train_dataset_workers: 32
num_hosts: 32
total_num_training_examples: 51,200,000
steps_per_epoch: 7798
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_nst_loss	train_loss	validation_nst_wer	validation_nst_cer	validation_nst_exact_wer	validation_nst_exact_cer	validation_clean_stortinget_no_loss	validation_clean_stortinget_no_wer	validation_clean_stortinget_no_cer	validation_clean_stortinget_no_exact_wer	validation_clean_stortinget_no_exact_cer
0	0.4271	0.9562	2.1721	0.6246	2.7056	0.7070	0.6866	8.5836	5.4517	11.4126	5.8853
5000	0.4400	0.5815	2.6621	0.7765	3.1629	0.8526	0.7085	9.1000	5.7626	12.1172	6.2354
10000	0.4377	0.5548	2.2701	0.6740	2.9016	0.7720	0.6845	9.2823	5.9461	12.1717	6.4073
15000	0.4332	0.5112	2.3246	0.6917	2.8799	0.7775	0.7101	9.1307	5.8030	11.9654	6.2408
20000	0.4345	0.5066	2.3518	0.7122	2.8962	0.7940	0.7083	9.0668	5.8133	11.9867	6.2755

Framework versions

Transformers 4.35.2
Datasets 2.15.0
Tokenizers 0.14.1