File size: 2,385 Bytes
6fde0cc
 
 
 
 
 
 
 
 
 
 
d264e9f
6fde0cc
 
 
 
 
d264e9f
6fde0cc
 
 
 
 
 
 
d264e9f
6fde0cc
d264e9f
6fde0cc
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
language:
- ga
license: apache-2.0
tags:
- whisper-event
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_11_0
metrics:
- wer
base_model: kpriyanshu256/whisper-large-v2-br-1000-32-1e-05
model-index:
- name: whisper-large-v2-Irish
  results:
  - task:
      type: automatic-speech-recognition
      name: Automatic Speech Recognition
    dataset:
      name: Common Voice 11.0
      type: mozilla-foundation/common_voice_11_0
      config: ga-IE
      split: test
      args: ga-IE
    metrics:
    - type: wer
      value: 42.36353077816493
      name: Wer
---

<!-- 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-large-v2-Irish

This model is a fine-tuned version of [kpriyanshu256/whisper-large-v2-br-1000-32-1e-05](https://huggingface.co/kpriyanshu256/whisper-large-v2-br-1000-32-1e-05) on the Common Voice 11.0 dataset.
It achieves the following results on the evaluation set:
- Loss: 1.2167
- Wer: 42.3635

## 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: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- training_steps: 800

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer     |
|:-------------:|:-----:|:----:|:---------------:|:-------:|
| 0.4381        | 3.0   | 100  | 0.9324          | 47.2416 |
| 0.0465        | 6.01  | 200  | 1.0565          | 45.6156 |
| 0.0169        | 9.02  | 300  | 1.0763          | 43.2636 |
| 0.0063        | 12.02 | 400  | 1.1362          | 44.0476 |
| 0.0024        | 15.03 | 500  | 1.1534          | 42.7410 |
| 0.0011        | 18.03 | 600  | 1.1959          | 42.3926 |
| 0.0009        | 21.04 | 700  | 1.2123          | 42.1893 |
| 0.0008        | 24.04 | 800  | 1.2167          | 42.3635 |


### Framework versions

- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2