shimonbh commited on
Commit
cb55e53
·
verified ·
1 Parent(s): 59bf11c

Model save

Browse files
README.md CHANGED
@@ -1,14 +1,29 @@
1
  ---
2
- base_model: openai/whisper-small
3
- datasets:
4
- - common_voice_17_0
5
  library_name: transformers
6
  license: apache-2.0
 
7
  tags:
8
  - generated_from_trainer
 
 
 
 
9
  model-index:
10
  - name: stt-small
11
- results: []
 
 
 
 
 
 
 
 
 
 
 
 
 
12
  ---
13
 
14
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -17,6 +32,9 @@ should probably proofread and complete it, then remove this comment. -->
17
  # stt-small
18
 
19
  This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the common_voice_17_0 dataset.
 
 
 
20
 
21
  ## Model description
22
 
@@ -45,6 +63,22 @@ The following hyperparameters were used during training:
45
  - training_steps: 4000
46
  - mixed_precision_training: Native AMP
47
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
48
  ### Framework versions
49
 
50
  - Transformers 4.44.2
 
1
  ---
 
 
 
2
  library_name: transformers
3
  license: apache-2.0
4
+ base_model: openai/whisper-small
5
  tags:
6
  - generated_from_trainer
7
+ datasets:
8
+ - common_voice_17_0
9
+ metrics:
10
+ - wer
11
  model-index:
12
  - name: stt-small
13
+ results:
14
+ - task:
15
+ name: Automatic Speech Recognition
16
+ type: automatic-speech-recognition
17
+ dataset:
18
+ name: common_voice_17_0
19
+ type: common_voice_17_0
20
+ config: he
21
+ split: test
22
+ args: he
23
+ metrics:
24
+ - name: Wer
25
+ type: wer
26
+ value: 71.81719260065287
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
32
  # stt-small
33
 
34
  This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the common_voice_17_0 dataset.
35
+ It achieves the following results on the evaluation set:
36
+ - Loss: 0.9486
37
+ - Wer: 71.8172
38
 
39
  ## Model description
40
 
 
63
  - training_steps: 4000
64
  - mixed_precision_training: Native AMP
65
 
66
+ ### Training results
67
+
68
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
69
+ |:-------------:|:-------:|:----:|:---------------:|:-------:|
70
+ | 0.1591 | 5.1948 | 400 | 0.6390 | 74.0479 |
71
+ | 0.0164 | 10.3896 | 800 | 0.7373 | 74.3199 |
72
+ | 0.0078 | 15.5844 | 1200 | 0.8321 | 73.3950 |
73
+ | 0.0028 | 20.7792 | 1600 | 0.8744 | 74.2111 |
74
+ | 0.0012 | 25.9740 | 2000 | 0.9024 | 71.8172 |
75
+ | 0.0003 | 31.1688 | 2400 | 0.9137 | 72.0348 |
76
+ | 0.0002 | 36.3636 | 2800 | 0.9279 | 71.8172 |
77
+ | 0.0002 | 41.5584 | 3200 | 0.9391 | 71.8172 |
78
+ | 0.0002 | 46.7532 | 3600 | 0.9457 | 71.8172 |
79
+ | 0.0002 | 51.9481 | 4000 | 0.9486 | 71.8172 |
80
+
81
+
82
  ### Framework versions
83
 
84
  - Transformers 4.44.2
runs/Jan05_20-51-17_698e1c90f7bd/events.out.tfevents.1736110281.698e1c90f7bd.40.1 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:db2b2621c09e6f70d1f61419ad8675958a135fcf0027151d8a8500216609bd1f
3
- size 10380
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99879784ec7b2a128f1d89096ab474b4a9e07c30e932ff7a702e5c8bbca75ab4
3
+ size 10734