File size: 1,750 Bytes
be52b84
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
---
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper_AN_demo
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# whisper_AN_demo

This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5163
- Wer Ortho: 34.7268
- Wer: 29.6857

## 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: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 50
- training_steps: 500

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer     |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:-------:|
| 0.6419        | 0.05  | 100  | 0.8019          | 31.6674   | 26.4281 |
| 0.2769        | 0.1   | 200  | 0.5559          | 32.3914   | 27.3457 |
| 0.2674        | 0.15  | 300  | 0.5354          | 39.1172   | 33.8151 |
| 0.2672        | 0.19  | 400  | 0.5247          | 34.6333   | 29.6628 |
| 0.2876        | 0.24  | 500  | 0.5163          | 34.7268   | 29.6857 |


### Framework versions

- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1