Hugging Face's logo

emilios
/
whisper-medium-el

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
Safetensors
Greek
whisper
hf-asr-leaderboard
whisper-medium
mozilla-foundation/common_voice_11_0
greek
whisper-event
Generated from Trainer
Eval Results (legacy)
Model card Files
xet
 Metrics Community
4
emilios commited on
Commit
4e3c18c
·
1 Parent(s): 3b55347

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +22 -18
README.md CHANGED
@@ -1,31 +1,35 @@
1
  ---
2
  language:
3
  - el
4
- license: apache-2.0
5
  tags:
6
- - whisper-medium
7
- - mozilla-foundation/common_voice_11_0
8
- - greek
9
- - whisper-event
10
- - hf-asr-leaderboard
11
  - generated_from_trainer
12
  datasets:
13
- - mozilla-foundation/common_voice_11_0
14
  model-index:
15
- - name: Whisper medium El - Greek One
16
  results: []
17
  ---
18
 
19
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
20
  should probably proofread and complete it, then remove this comment. -->
21
 
22
- # Whisper medium El - Greek One
23
 
24
- This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the Common Voice 11.0 dataset.
 
 
 
 
 
 
 
 
25
 
26
  ## Model description
27
 
28
- This is a fine-tuned version of the medium whisper model for Greek language.
29
 
30
  ## Intended uses & limitations
31
 
@@ -41,20 +45,20 @@ More information needed
41
 
42
  The following hyperparameters were used during training:
43
  - learning_rate: 1e-05
44
- - train_batch_size: 4
45
  - eval_batch_size: 8
46
  - seed: 42
47
- - gradient_accumulation_steps: 4
48
- - total_train_batch_size: 16
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: linear
51
  - lr_scheduler_warmup_steps: 500
52
- - training_steps: 4000
53
  - mixed_precision_training: Native AMP
54
 
55
  ### Framework versions
56
 
57
- - Transformers 4.25.0.dev0
58
- - Pytorch 1.13.0
59
- - Datasets 2.6.1
60
  - Tokenizers 0.13.2
 
1
  ---
2
  language:
3
  - el
 
4
  tags:
5
+ - hf-asr-leaderboard, whisper-large, mozilla-foundation/common_voice_11_0, greek,
6
+ whisper-event
 
 
 
7
  - generated_from_trainer
8
  datasets:
9
+ - mozilla-foundation/common_voice_11_0, google/fleurs
10
  model-index:
11
+ - name: Whisper Medium El - Greek One
12
  results: []
13
  ---
14
 
15
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
  should probably proofread and complete it, then remove this comment. -->
17
 
18
+ # Whisper Medium El - Greek One
19
 
20
+ This model is a fine-tuned version of [openai/medium-small](https://huggingface.co/openai/medium-small) on the Common Voice 11.0 dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - eval_loss: 0.3928
23
+ - eval_wer: 14.8031
24
+ - eval_runtime: 1229.0439
25
+ - eval_samples_per_second: 1.38
26
+ - eval_steps_per_second: 0.172
27
+ - epoch: 20.0
28
+ - step: 1000
29
 
30
  ## Model description
31
 
32
+ More information needed
33
 
34
  ## Intended uses & limitations
35
 
 
45
 
46
  The following hyperparameters were used during training:
47
  - learning_rate: 1e-05
48
+ - train_batch_size: 20
49
  - eval_batch_size: 8
50
  - seed: 42
51
+ - gradient_accumulation_steps: 2
52
+ - total_train_batch_size: 40
53
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
54
  - lr_scheduler_type: linear
55
  - lr_scheduler_warmup_steps: 500
56
+ - training_steps: 12000
57
  - mixed_precision_training: Native AMP
58
 
59
  ### Framework versions
60
 
61
+ - Transformers 4.26.0.dev0
62
+ - Pytorch 1.13.0+cu116
63
+ - Datasets 2.7.1.dev0
64
  - Tokenizers 0.13.2