haseong8012
commited on
Commit
•
aab1248
1
Parent(s):
ceda41a
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- ko
|
4 |
+
license: apache-2.0
|
5 |
+
tags:
|
6 |
+
- hf-asr-leaderboard
|
7 |
+
- generated_from_trainer
|
8 |
+
datasets:
|
9 |
+
- haseong8012/korean-child-command-voice_train-0-10000_smaplingRate-16000
|
10 |
+
metrics:
|
11 |
+
- wer
|
12 |
+
model-index:
|
13 |
+
- name: openai/whisper-small-Ko-haseong8012/korean-child-command-voice_train-0-10000_smaplingRate-16000
|
14 |
+
results:
|
15 |
+
- task:
|
16 |
+
name: Automatic Speech Recognition
|
17 |
+
type: automatic-speech-recognition
|
18 |
+
dataset:
|
19 |
+
name: haseong8012/korean-child-command-voice_train-0-10000_smaplingRate-16000
|
20 |
+
type: haseong8012/korean-child-command-voice_train-0-10000_smaplingRate-16000
|
21 |
+
args: 'config: ko, split: train, validation, test'
|
22 |
+
metrics:
|
23 |
+
- name: Wer
|
24 |
+
type: wer
|
25 |
+
value: 14.96458761708933
|
26 |
+
---
|
27 |
+
|
28 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
29 |
+
should probably proofread and complete it, then remove this comment. -->
|
30 |
+
|
31 |
+
# openai/whisper-small-Ko-haseong8012/korean-child-command-voice_train-0-10000_smaplingRate-16000
|
32 |
+
|
33 |
+
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the haseong8012/korean-child-command-voice_train-0-10000_smaplingRate-16000 dataset.
|
34 |
+
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 0.1613
|
36 |
+
- Wer: 14.9646
|
37 |
+
- Cer: 7.4814
|
38 |
+
|
39 |
+
## Model description
|
40 |
+
|
41 |
+
More information needed
|
42 |
+
|
43 |
+
## Intended uses & limitations
|
44 |
+
|
45 |
+
More information needed
|
46 |
+
|
47 |
+
## Training and evaluation data
|
48 |
+
|
49 |
+
More information needed
|
50 |
+
|
51 |
+
## Training procedure
|
52 |
+
|
53 |
+
### Training hyperparameters
|
54 |
+
|
55 |
+
The following hyperparameters were used during training:
|
56 |
+
- learning_rate: 1.25e-05
|
57 |
+
- train_batch_size: 32
|
58 |
+
- eval_batch_size: 16
|
59 |
+
- seed: 42
|
60 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
61 |
+
- lr_scheduler_type: linear
|
62 |
+
- lr_scheduler_warmup_steps: 500
|
63 |
+
- training_steps: 5000
|
64 |
+
- mixed_precision_training: Native AMP
|
65 |
+
|
66 |
+
### Training results
|
67 |
+
|
68 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|
69 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|:------:|
|
70 |
+
| 0.0098 | 4.0 | 1000 | 0.1831 | 18.0032 | 9.0777 |
|
71 |
+
| 0.0007 | 8.0 | 2000 | 0.1634 | 15.6271 | 7.7868 |
|
72 |
+
| 0.0002 | 12.0 | 3000 | 0.1611 | 15.2159 | 7.5300 |
|
73 |
+
| 0.0001 | 16.0 | 4000 | 0.1605 | 15.0331 | 7.5370 |
|
74 |
+
| 0.0001 | 20.0 | 5000 | 0.1613 | 14.9646 | 7.4814 |
|
75 |
+
|
76 |
+
|
77 |
+
### Framework versions
|
78 |
+
|
79 |
+
- Transformers 4.28.0
|
80 |
+
- Pytorch 2.1.0+cu121
|
81 |
+
- Datasets 2.14.5
|
82 |
+
- Tokenizers 0.13.3
|