Update stats.md
Browse files
stats.md
ADDED
|
@@ -0,0 +1,96 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- 'no'
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
base_model: NbAiLab/nb-whisper-tiny-RC1
|
| 6 |
+
tags:
|
| 7 |
+
- audio
|
| 8 |
+
- asr
|
| 9 |
+
- automatic-speech-recognition
|
| 10 |
+
- hf-asr-leaderboard
|
| 11 |
+
model-index:
|
| 12 |
+
- name: nb-whisper-tiny-v0.8-vad3
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information Keras had access to. You should
|
| 17 |
+
probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# nb-whisper-tiny-v0.8-vad3
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [NbAiLab/nb-whisper-tiny-RC1](https://huggingface.co/NbAiLab/nb-whisper-tiny-RC1) on the NbAiLab/ncc_speech_styling_v2_vad3 dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- step: 49999
|
| 24 |
+
- validation_nst_loss: 0.5107
|
| 25 |
+
- train_loss: 0.6596
|
| 26 |
+
- validation_nst_wer: 7.8121
|
| 27 |
+
- validation_nst_cer: 2.4815
|
| 28 |
+
- validation_nst_exact_wer: 8.6559
|
| 29 |
+
- validation_nst_exact_cer: 2.6081
|
| 30 |
+
- validation_clean_stortinget_no_loss: 0.7180
|
| 31 |
+
- validation_clean_stortinget_no_wer: 16.0137
|
| 32 |
+
- validation_clean_stortinget_no_cer: 8.6585
|
| 33 |
+
- validation_clean_stortinget_no_exact_wer: 19.8672
|
| 34 |
+
- validation_clean_stortinget_no_exact_cer: 9.3568
|
| 35 |
+
|
| 36 |
+
## Model description
|
| 37 |
+
|
| 38 |
+
More information needed
|
| 39 |
+
|
| 40 |
+
## Intended uses & limitations
|
| 41 |
+
|
| 42 |
+
More information needed
|
| 43 |
+
|
| 44 |
+
## Training and evaluation data
|
| 45 |
+
|
| 46 |
+
More information needed
|
| 47 |
+
|
| 48 |
+
## Training procedure
|
| 49 |
+
|
| 50 |
+
### Training hyperparameters
|
| 51 |
+
|
| 52 |
+
The following hyperparameters were used during training:
|
| 53 |
+
- learning_rate: 0.00015
|
| 54 |
+
- lr_scheduler_type: linear
|
| 55 |
+
- per_device_train_batch_size: 32
|
| 56 |
+
- total_train_batch_size_per_node: 128
|
| 57 |
+
- total_train_batch_size: 1024
|
| 58 |
+
- total_optimization_steps: 50,000
|
| 59 |
+
- starting_optimization_step: None
|
| 60 |
+
- finishing_optimization_step: 50,000
|
| 61 |
+
- num_train_dataset_workers: 32
|
| 62 |
+
- num_hosts: 8
|
| 63 |
+
- total_num_training_examples: 51,200,000
|
| 64 |
+
- steps_per_epoch: 7455
|
| 65 |
+
- num_beams: None
|
| 66 |
+
- weight_decay: 0.01
|
| 67 |
+
- adam_beta1: 0.9
|
| 68 |
+
- adam_beta2: 0.98
|
| 69 |
+
- adam_epsilon: 1e-06
|
| 70 |
+
- dropout: True
|
| 71 |
+
- bpe_dropout_probability: 0.2
|
| 72 |
+
- activation_dropout_probability: 0.1
|
| 73 |
+
|
| 74 |
+
### Training results
|
| 75 |
+
|
| 76 |
+
| 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 |
|
| 77 |
+
|:-----:|:-------------------:|:----------:|:------------------:|:------------------:|:------------------------:|:------------------------:|:-----------------------------------:|:----------------------------------:|:----------------------------------:|:----------------------------------------:|:----------------------------------------:|
|
| 78 |
+
| 0 | 0.5115 | 1.2809 | 7.9808 | 2.5868 | 8.9172 | 2.7299 | 0.6754 | 16.0516 | 8.8512 | 19.8245 | 9.5377 |
|
| 79 |
+
| 5000 | 0.5513 | 0.7948 | 9.8699 | 3.1480 | 10.7083 | 3.2666 | 0.8315 | 18.2473 | 9.7512 | 22.3081 | 10.4913 |
|
| 80 |
+
| 10000 | 0.5434 | 0.7729 | 9.5433 | 3.0613 | 10.3490 | 3.1851 | 0.7654 | 17.9536 | 9.6422 | 21.9855 | 10.3974 |
|
| 81 |
+
| 15000 | 0.5380 | 0.7629 | 9.0914 | 2.9523 | 9.9243 | 3.0779 | 0.7651 | 17.6694 | 9.2759 | 21.5443 | 9.9923 |
|
| 82 |
+
| 20000 | 0.5364 | 0.7173 | 8.7974 | 2.8684 | 9.6412 | 2.9882 | 0.7469 | 17.0725 | 9.1391 | 20.9963 | 9.8515 |
|
| 83 |
+
| 25000 | 0.5311 | 0.7168 | 8.6559 | 2.7453 | 9.5160 | 2.8700 | 0.7347 | 16.8167 | 9.0130 | 20.7615 | 9.7248 |
|
| 84 |
+
| 30000 | 0.5282 | 0.6977 | 8.4545 | 2.7090 | 9.2711 | 2.8362 | 0.7588 | 16.5632 | 8.9523 | 20.4650 | 9.6476 |
|
| 85 |
+
| 35000 | 0.5232 | 0.6851 | 8.4000 | 2.7062 | 9.2547 | 2.8316 | 0.7314 | 16.2577 | 8.7608 | 20.2135 | 9.4749 |
|
| 86 |
+
| 40000 | 0.5215 | 0.6878 | 8.3238 | 2.6251 | 9.1676 | 2.7528 | 0.7387 | 16.3098 | 8.8231 | 20.1542 | 9.5131 |
|
| 87 |
+
| 45000 | 0.5156 | 0.6901 | 7.9264 | 2.5309 | 8.7158 | 2.6466 | 0.7236 | 16.1037 | 8.8698 | 19.9620 | 9.5607 |
|
| 88 |
+
| 49999 | 0.5107 | 0.6596 | 7.8121 | 2.4815 | 8.6559 | 2.6081 |
|
| 89 |
+
| 49999 | 0.7180 | 0.6596 | 16.0137 | 8.6585 | 19.8672 | 9.3568 |
|
| 90 |
+
|
| 91 |
+
|
| 92 |
+
### Framework versions
|
| 93 |
+
|
| 94 |
+
- Transformers 4.34.1
|
| 95 |
+
- Datasets 2.16.1
|
| 96 |
+
- Tokenizers 0.14.1
|