thiagobarbosa commited on
Commit
74ec353
1 Parent(s): 9654ec4

Model save

Browse files
Files changed (1) hide show
  1. README.md +20 -23
README.md CHANGED
@@ -1,43 +1,40 @@
1
  ---
2
- language:
3
- - pt
4
  license: apache-2.0
5
  base_model: openai/whisper-base
6
  tags:
7
- - hf-asr-leaderboard
8
  - generated_from_trainer
9
  datasets:
10
- - mozilla-foundation/common_voice_16_0
11
  metrics:
12
  - wer
13
  model-index:
14
- - name: Whisper Base using Common Voice 16 (pt)
15
  results:
16
  - task:
17
  name: Automatic Speech Recognition
18
  type: automatic-speech-recognition
19
  dataset:
20
- name: Mozilla Common Voices - 16.0 - Portuguese
21
- type: mozilla-foundation/common_voice_16_0
22
  config: pt
23
- split: test[0:5400]
24
  args: pt
25
  metrics:
26
  - name: Wer
27
  type: wer
28
- value: 25.542580301884676
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 using Common Voice 16 (pt)
35
 
36
- This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the Mozilla Common Voices - 16.0 - Portuguese dataset.
37
  It achieves the following results on the evaluation set:
38
- - Loss: 0.3952
39
- - Wer: 25.5426
40
- - Wer Normalized: 19.7098
41
 
42
  ## Model description
43
 
@@ -56,7 +53,7 @@ More information needed
56
  ### Training hyperparameters
57
 
58
  The following hyperparameters were used during training:
59
- - learning_rate: 5e-05
60
  - train_batch_size: 16
61
  - eval_batch_size: 8
62
  - seed: 42
@@ -70,14 +67,14 @@ The following hyperparameters were used during training:
70
 
71
  | Training Loss | Epoch | Step | Validation Loss | Wer | Wer Normalized |
72
  |:-------------:|:-----:|:----:|:---------------:|:-------:|:--------------:|
73
- | 0.5344 | 0.37 | 500 | 0.5264 | 35.9965 | 30.1234 |
74
- | 0.438 | 0.74 | 1000 | 0.4904 | 33.4453 | 28.3776 |
75
- | 0.1871 | 1.11 | 1500 | 0.4595 | 30.3929 | 24.5163 |
76
- | 0.1955 | 1.48 | 2000 | 0.4342 | 28.6566 | 22.9762 |
77
- | 0.1754 | 1.85 | 2500 | 0.4199 | 28.2674 | 22.4147 |
78
- | 0.0649 | 2.22 | 3000 | 0.4090 | 26.7860 | 20.7689 |
79
- | 0.0595 | 2.59 | 3500 | 0.4026 | 26.1839 | 20.2018 |
80
- | 0.0626 | 2.96 | 4000 | 0.3952 | 25.5426 | 19.7098 |
81
 
82
 
83
  ### Framework versions
 
1
  ---
 
 
2
  license: apache-2.0
3
  base_model: openai/whisper-base
4
  tags:
 
5
  - generated_from_trainer
6
  datasets:
7
+ - common_voice_16_0
8
  metrics:
9
  - wer
10
  model-index:
11
+ - name: whisper-base-common-voice-16-pt-v6
12
  results:
13
  - task:
14
  name: Automatic Speech Recognition
15
  type: automatic-speech-recognition
16
  dataset:
17
+ name: common_voice_16_0
18
+ type: common_voice_16_0
19
  config: pt
20
+ split: test
21
  args: pt
22
  metrics:
23
  - name: Wer
24
  type: wer
25
+ value: 25.436328377504847
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
29
  should probably proofread and complete it, then remove this comment. -->
30
 
31
+ # whisper-base-common-voice-16-pt-v6
32
 
33
+ This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the common_voice_16_0 dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 0.3552
36
+ - Wer: 25.4363
37
+ - Wer Normalized: 19.4668
38
 
39
  ## Model description
40
 
 
53
  ### Training hyperparameters
54
 
55
  The following hyperparameters were used during training:
56
+ - learning_rate: 2e-05
57
  - train_batch_size: 16
58
  - eval_batch_size: 8
59
  - seed: 42
 
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Wer | Wer Normalized |
69
  |:-------------:|:-----:|:----:|:---------------:|:-------:|:--------------:|
70
+ | 0.6085 | 0.19 | 500 | 0.4465 | 32.1833 | 25.3383 |
71
+ | 0.4624 | 0.37 | 1000 | 0.4131 | 28.9867 | 22.8488 |
72
+ | 0.4375 | 0.56 | 1500 | 0.3936 | 27.8135 | 21.3817 |
73
+ | 0.4372 | 0.74 | 2000 | 0.3784 | 27.5695 | 21.7171 |
74
+ | 0.4704 | 0.93 | 2500 | 0.3630 | 26.1167 | 20.5133 |
75
+ | 0.2013 | 1.11 | 3000 | 0.3600 | 25.5462 | 19.7750 |
76
+ | 0.2261 | 1.3 | 3500 | 0.3570 | 25.5010 | 19.5181 |
77
+ | 0.2118 | 1.48 | 4000 | 0.3552 | 25.4363 | 19.4668 |
78
 
79
 
80
  ### Framework versions