File size: 2,528 Bytes
c4cf11e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
---
library_name: transformers
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
datasets:
- PolyAI/minds14
metrics:
- wer
model-index:
- name: whisper-tiny-en-US
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: PolyAI/minds14
      type: PolyAI/minds14
      config: en-US
      split: train[450:]
      args: en-US
    metrics:
    - name: Wer
      type: wer
      value: 0.34887839433293977
---

<!-- 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-tiny-en-US

This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the PolyAI/minds14 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6638
- Wer Ortho: 34.5466
- Wer: 0.3489

## 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
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch   | Step | Validation Loss | Wer Ortho | Wer    |
|:-------------:|:-------:|:----:|:---------------:|:---------:|:------:|
| 0.7657        | 1.7857  | 50   | 0.5870          | 39.4818   | 0.3932 |
| 0.2562        | 3.5714  | 100  | 0.4866          | 34.8550   | 0.3483 |
| 0.0666        | 5.3571  | 150  | 0.5190          | 34.5466   | 0.3489 |
| 0.0228        | 7.1429  | 200  | 0.5649          | 32.4491   | 0.3288 |
| 0.0065        | 8.9286  | 250  | 0.5845          | 32.0173   | 0.3229 |
| 0.0018        | 10.7143 | 300  | 0.6142          | 33.6212   | 0.3400 |
| 0.0012        | 12.5    | 350  | 0.6320          | 33.3128   | 0.3371 |
| 0.0008        | 14.2857 | 400  | 0.6443          | 34.1764   | 0.3465 |
| 0.0007        | 16.0714 | 450  | 0.6548          | 34.2381   | 0.3447 |
| 0.0007        | 17.8571 | 500  | 0.6638          | 34.5466   | 0.3489 |


### Framework versions

- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.0
- Tokenizers 0.19.1