Update stats.md
Browse files
stats.md
ADDED
@@ -0,0 +1,93 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- 'no'
|
4 |
+
license: apache-2.0
|
5 |
+
base_model: NbAiLabBeta/nb-whisper-medium
|
6 |
+
tags:
|
7 |
+
- audio
|
8 |
+
- asr
|
9 |
+
- automatic-speech-recognition
|
10 |
+
- hf-asr-leaderboard
|
11 |
+
model-index:
|
12 |
+
- name: nb-whisper-medium-v0.7-semantic
|
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-medium-v0.7-semantic
|
20 |
+
|
21 |
+
This model is a fine-tuned version of [NbAiLabBeta/nb-whisper-medium](https://huggingface.co/NbAiLabBeta/nb-whisper-medium) on the NbAiLab/ncc_speech_styling_v4 dataset.
|
22 |
+
It achieves the following results on the evaluation set:
|
23 |
+
- step: 249
|
24 |
+
- validation_nst_loss: 0.6407
|
25 |
+
- train_loss: 0.6265
|
26 |
+
- validation_nst_wer: 2.4389
|
27 |
+
- validation_nst_cer: 0.7737
|
28 |
+
- validation_nst_exact_wer: 3.1412
|
29 |
+
- validation_nst_exact_cer: 0.8810
|
30 |
+
- validation_clean_stortinget_no_loss: 0.7441
|
31 |
+
- validation_clean_stortinget_no_wer: 9.1639
|
32 |
+
- validation_clean_stortinget_no_cer: 5.9596
|
33 |
+
- validation_clean_stortinget_no_exact_wer: 12.2050
|
34 |
+
- validation_clean_stortinget_no_exact_cer: 6.4346
|
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: 2.5e-05
|
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: 250
|
59 |
+
- starting_optimization_step: None
|
60 |
+
- finishing_optimization_step: 250
|
61 |
+
- num_train_dataset_workers: 32
|
62 |
+
- num_hosts: 8
|
63 |
+
- total_num_training_examples: 256,000
|
64 |
+
- steps_per_epoch: _To be computed after first epoch_
|
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 | 0.4372 | 1.1581 | 2.2865 | 0.7038 | 2.9452 | 0.8059 | 0.7951 | 8.8939 | 5.6567 | 11.9393 | 6.1348 |
|
79 |
+
| 40 | 0.7551 | 0.7309 | 2.7546 | 0.9443 | 3.4950 | 1.0568 | 0.7489 | 9.8437 | 6.4691 | 12.9830 | 6.9559 |
|
80 |
+
| 80 | 0.7655 | 0.6341 | 2.5695 | 0.8250 | 3.3317 | 0.9414 | 0.7355 | 9.3794 | 6.1000 | 12.4090 | 6.5664 |
|
81 |
+
| 120 | 0.7493 | 0.6262 | 2.4716 | 0.8129 | 3.2283 | 0.9286 | 0.7432 | 9.3344 | 6.0544 | 12.3592 | 6.5247 |
|
82 |
+
| 160 | 0.7165 | 0.6353 | 2.5859 | 0.8259 | 3.3154 | 0.9378 | 0.7455 | 9.3179 | 6.0647 | 12.3852 | 6.5418 |
|
83 |
+
| 200 | 0.6678 | 0.6169 | 2.4661 | 0.7812 | 3.1575 | 0.8892 | 0.7430 | 9.2752 | 6.0298 | 12.3378 | 6.5079 |
|
84 |
+
| 240 | 0.6409 | 0.6385 | 2.4498 | 0.7849 | 3.1520 | 0.8911 | 0.7434 | 9.1900 | 5.9941 | 12.1931 | 6.4639 |
|
85 |
+
| 249 | 0.6407 | 0.6265 | 2.4389 | 0.7737 | 3.1412 | 0.8810 |
|
86 |
+
| 249 | 0.7441 | 0.6265 | 9.1639 | 5.9596 | 12.2050 | 6.4346 |
|
87 |
+
|
88 |
+
|
89 |
+
### Framework versions
|
90 |
+
|
91 |
+
- Transformers 4.34.1
|
92 |
+
- Datasets 2.15.0
|
93 |
+
- Tokenizers 0.14.1
|