update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- hy
|
4 |
+
license: apache-2.0
|
5 |
+
tags:
|
6 |
+
- whisper-event
|
7 |
+
- generated_from_trainer
|
8 |
+
datasets:
|
9 |
+
- mozilla-foundation/common_voice_11_0
|
10 |
+
- google/fleurs
|
11 |
+
metrics:
|
12 |
+
- wer
|
13 |
+
model-index:
|
14 |
+
- name: whisper-base-hy
|
15 |
+
results:
|
16 |
+
- task:
|
17 |
+
name: Automatic Speech Recognition
|
18 |
+
type: automatic-speech-recognition
|
19 |
+
dataset:
|
20 |
+
name: Common Voice 11.0
|
21 |
+
type: mozilla-foundation/common_voice_11_0
|
22 |
+
config: hy-AM
|
23 |
+
split: test
|
24 |
+
args: hy-AM
|
25 |
+
metrics:
|
26 |
+
- name: Wer
|
27 |
+
type: wer
|
28 |
+
value: 22.36842105263158
|
29 |
+
---
|
30 |
+
|
31 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
32 |
+
should probably proofread and complete it, then remove this comment. -->
|
33 |
+
|
34 |
+
# whisper-base-hy
|
35 |
+
|
36 |
+
This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the Common Voice 11.0 dataset.
|
37 |
+
It achieves the following results on the evaluation set:
|
38 |
+
- Loss: 0.2204
|
39 |
+
- Wer: 22.3684
|
40 |
+
|
41 |
+
## Model description
|
42 |
+
|
43 |
+
More information needed
|
44 |
+
|
45 |
+
## Intended uses & limitations
|
46 |
+
|
47 |
+
More information needed
|
48 |
+
|
49 |
+
## Training and evaluation data
|
50 |
+
|
51 |
+
More information needed
|
52 |
+
|
53 |
+
## Training procedure
|
54 |
+
|
55 |
+
### Training hyperparameters
|
56 |
+
|
57 |
+
The following hyperparameters were used during training:
|
58 |
+
- learning_rate: 1e-05
|
59 |
+
- train_batch_size: 4
|
60 |
+
- eval_batch_size: 4
|
61 |
+
- seed: 42
|
62 |
+
- gradient_accumulation_steps: 16
|
63 |
+
- total_train_batch_size: 64
|
64 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
65 |
+
- lr_scheduler_type: linear
|
66 |
+
- lr_scheduler_warmup_steps: 200
|
67 |
+
- training_steps: 2000
|
68 |
+
- mixed_precision_training: Native AMP
|
69 |
+
|
70 |
+
### Training results
|
71 |
+
|
72 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
73 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|
|
74 |
+
| 0.1394 | 5.87 | 400 | 0.1780 | 28.2895 |
|
75 |
+
| 0.0536 | 11.75 | 800 | 0.1739 | 24.6053 |
|
76 |
+
| 0.0247 | 17.64 | 1200 | 0.2098 | 22.9605 |
|
77 |
+
| 0.0154 | 23.52 | 1600 | 0.2035 | 22.1382 |
|
78 |
+
| 0.0103 | 29.41 | 2000 | 0.2204 | 22.3684 |
|
79 |
+
|
80 |
+
|
81 |
+
### Framework versions
|
82 |
+
|
83 |
+
- Transformers 4.26.0.dev0
|
84 |
+
- Pytorch 1.13.0+cu117
|
85 |
+
- Datasets 2.7.1.dev0
|
86 |
+
- Tokenizers 0.13.2
|