File size: 2,531 Bytes
db96f8b
14da60c
 
db96f8b
 
14da60c
db96f8b
 
14da60c
db96f8b
 
 
14da60c
db96f8b
 
 
 
 
14da60c
 
db96f8b
 
 
 
 
 
14da60c
db96f8b
 
 
 
 
14da60c
db96f8b
14da60c
db96f8b
d1eddd9
14da60c
db96f8b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c044832
db96f8b
 
 
6a213a7
 
 
 
 
 
 
 
 
 
 
 
c044832
db96f8b
 
 
 
 
 
 
 
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
90
---
language:
- el
license: apache-2.0
tags:
- whisper-event
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_11_0
metrics:
- wer
model-index:
- name: Whisper medium Greek El Greco
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: mozilla-foundation/common_voice_11_0 el
      type: mozilla-foundation/common_voice_11_0
      config: el
      split: test
      args: el
    metrics:
    - name: Wer
      type: wer
      value: 9.899702823179792
---

<!-- 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 medium Greek El Greco

This model is a fine-tuned version of [emilios/whisper-medium-el-n2](https://huggingface.co/emilios/whisper-medium-el-n2) on the mozilla-foundation/common_voice_11_0 el dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5669
- Wer: 9.8997

## 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: 3e-06
- train_batch_size: 32
- eval_batch_size: 16
- seed: 42
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 11000

### Training results

| Training Loss | Epoch  | Step  | Validation Loss | Wer     |
|:-------------:|:------:|:-----:|:---------------:|:-------:|
| 0.0014        | 58.82  | 1000  | 0.4951          | 10.3640 |
| 0.0006        | 117.65 | 2000  | 0.5181          | 10.2805 |
| 0.0007        | 175.82 | 3000  | 0.5317          | 10.1133 |
| 0.0004        | 234.65 | 4000  | 0.5396          | 10.1226 |
| 0.0004        | 293.47 | 5000  | 0.5532          | 10.1040 |
| 0.0013        | 352.29 | 6000  | 0.5645          | 10.0854 |
| 0.0002        | 411.12 | 7000  | 0.5669          | 10.1133 |
| 0.0001        | 469.94 | 8000  | 0.5669          | 9.8997  |
| 0.0001        | 528.76 | 9000  | 0.5645          | 9.9276  |
| 0.0001        | 587.82 | 10000 | 0.5674          | 9.9647  |
| 0.0003        | 646.82 | 11000 | 0.5669          | 9.9461  |


### Framework versions

- Transformers 4.26.0.dev0
- Pytorch 2.0.0.dev20221216+cu116
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2