Hugging Face's logo

cantillation
/
whisper-medium-he-teamim-silsuless

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Hebrew
whisper
hf-asr-leaderboard
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
cantillation commited on
Commit
12a1cd3
·
verified ·
1 Parent(s): 0f8ebb9

Training in progress, step 1500

Browse files
Files changed (8) hide show
  1. README.md +34 -34
  2. config.json +1 -1
  3. generation_config.json +1 -1
  4. model.safetensors +1 -1
  5. runs/Jan29_21-18-20_DESKTOP-N7AL96B/events.out.tfevents.1706555902.DESKTOP-N7AL96B +3 -0
  6. runs/Jan29_22-54-50_DESKTOP-N7AL96B/events.out.tfevents.1706561691.DESKTOP-N7AL96B +3 -0
  7. runs/Jan30_20-30-34_DESKTOP-N7AL96B/events.out.tfevents.1706639435.DESKTOP-N7AL96B +3 -0
  8. training_args.bin +2 -2
README.md CHANGED
@@ -20,8 +20,8 @@ should probably proofread and complete it, then remove this comment. -->
20
 
21
  This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on an unknown dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 0.1223
24
- - Wer: 34.4444
25
 
26
  ## Model description
27
 
@@ -53,41 +53,41 @@ The following hyperparameters were used during training:
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Wer |
55
  |:-------------:|:-----:|:----:|:---------------:|:-------:|
56
- | 0.0255 | 0.06 | 50 | 0.1183 | 88.8889 |
57
- | 0.0334 | 0.11 | 100 | 0.1405 | 100.0 |
58
- | 0.033 | 0.17 | 150 | 0.1348 | 100.0 |
59
- | 0.0273 | 0.23 | 200 | 0.1044 | 74.4444 |
60
- | 0.0252 | 0.28 | 250 | 0.1197 | 51.1111 |
61
- | 0.023 | 0.34 | 300 | 0.1047 | 100.0 |
62
- | 0.0188 | 0.4 | 350 | 0.1087 | 87.7778 |
63
- | 0.0195 | 0.45 | 400 | 0.1228 | 100.0 |
64
- | 0.0185 | 0.51 | 450 | 0.1469 | 100.0 |
65
- | 0.0164 | 0.57 | 500 | 0.1042 | 100.0 |
66
- | 0.0149 | 0.62 | 550 | 0.0898 | 100.0 |
67
- | 0.0113 | 0.68 | 600 | 0.0966 | 100.0 |
68
- | 0.0116 | 0.74 | 650 | 0.1017 | 100.0 |
69
- | 0.0124 | 0.8 | 700 | 0.0996 | 100.0 |
70
- | 0.0096 | 0.85 | 750 | 0.1081 | 87.7778 |
71
- | 0.0115 | 0.91 | 800 | 0.1050 | 21.1111 |
72
- | 0.0117 | 0.97 | 850 | 0.1121 | 87.7778 |
73
- | 0.0091 | 1.02 | 900 | 0.1258 | 88.8889 |
74
- | 0.0034 | 1.08 | 950 | 0.1386 | 74.4444 |
75
- | 0.004 | 1.14 | 1000 | 0.1354 | 60.0 |
76
- | 0.0031 | 1.19 | 1050 | 0.1392 | 58.8889 |
77
- | 0.0028 | 1.25 | 1100 | 0.1235 | 57.7778 |
78
- | 0.0029 | 1.31 | 1150 | 0.1253 | 57.7778 |
79
- | 0.0033 | 1.36 | 1200 | 0.1154 | 57.7778 |
80
- | 0.0021 | 1.42 | 1250 | 0.1206 | 35.5556 |
81
- | 0.0026 | 1.48 | 1300 | 0.1244 | 35.5556 |
82
- | 0.0026 | 1.53 | 1350 | 0.1223 | 34.4444 |
83
- | 0.0027 | 1.59 | 1400 | 0.1212 | 45.5556 |
84
- | 0.0024 | 1.65 | 1450 | 0.1219 | 35.5556 |
85
- | 0.0023 | 1.7 | 1500 | 0.1223 | 34.4444 |
86
 
87
 
88
  ### Framework versions
89
 
90
- - Transformers 4.36.2
91
- - Pytorch 1.13.1+cu117
92
  - Datasets 2.16.1
93
  - Tokenizers 0.15.0
 
20
 
21
  This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on an unknown dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 1.2972
24
+ - Wer: 48.1928
25
 
26
  ## Model description
27
 
 
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Wer |
55
  |:-------------:|:-----:|:----:|:---------------:|:-------:|
56
+ | 1.3296 | 0.01 | 50 | 1.5188 | 59.0361 |
57
+ | 0.1103 | 0.03 | 100 | 1.4528 | 50.6024 |
58
+ | 0.0956 | 0.04 | 150 | 1.4441 | 54.2169 |
59
+ | 0.0945 | 0.06 | 200 | 1.5300 | 60.2410 |
60
+ | 0.0797 | 0.07 | 250 | 1.3702 | 42.1687 |
61
+ | 0.0709 | 0.09 | 300 | 1.2879 | 48.1928 |
62
+ | 0.0612 | 0.1 | 350 | 1.3409 | 56.6265 |
63
+ | 0.0657 | 0.12 | 400 | 1.2878 | 51.8072 |
64
+ | 0.0651 | 0.13 | 450 | 1.2181 | 56.6265 |
65
+ | 0.0542 | 0.15 | 500 | 1.4363 | 48.1928 |
66
+ | 0.0543 | 0.16 | 550 | 1.5288 | 60.2410 |
67
+ | 0.0503 | 0.18 | 600 | 1.3793 | 54.2169 |
68
+ | 0.0499 | 0.19 | 650 | 1.3918 | 57.8313 |
69
+ | 0.0417 | 0.21 | 700 | 1.2606 | 45.7831 |
70
+ | 0.0425 | 0.22 | 750 | 1.2568 | 53.0120 |
71
+ | 0.0422 | 0.24 | 800 | 1.3723 | 55.4217 |
72
+ | 0.0374 | 0.25 | 850 | 1.3720 | 54.2169 |
73
+ | 0.0287 | 0.27 | 900 | 1.3009 | 50.6024 |
74
+ | 0.0272 | 0.28 | 950 | 1.3992 | 54.2169 |
75
+ | 0.0339 | 0.29 | 1000 | 1.2794 | 53.0120 |
76
+ | 0.0251 | 0.31 | 1050 | 1.3348 | 56.6265 |
77
+ | 0.0275 | 0.32 | 1100 | 1.2444 | 46.9880 |
78
+ | 0.0247 | 0.34 | 1150 | 1.2549 | 45.7831 |
79
+ | 0.0231 | 0.35 | 1200 | 1.2892 | 43.3735 |
80
+ | 0.0303 | 0.37 | 1250 | 1.3194 | 46.9880 |
81
+ | 0.0186 | 0.38 | 1300 | 1.2910 | 46.9880 |
82
+ | 0.0188 | 0.4 | 1350 | 1.2911 | 45.7831 |
83
+ | 0.0132 | 0.41 | 1400 | 1.3061 | 48.1928 |
84
+ | 0.0173 | 0.43 | 1450 | 1.3096 | 48.1928 |
85
+ | 0.0153 | 0.44 | 1500 | 1.2972 | 48.1928 |
86
 
87
 
88
  ### Framework versions
89
 
90
+ - Transformers 4.37.0.dev0
91
+ - Pytorch 2.1.2
92
  - Datasets 2.16.1
93
  - Tokenizers 0.15.0
config.json CHANGED
@@ -46,7 +46,7 @@
46
  "scale_embedding": false,
47
  "suppress_tokens": [],
48
  "torch_dtype": "float32",
49
- "transformers_version": "4.36.2",
50
  "use_cache": true,
51
  "use_weighted_layer_sum": false,
52
  "vocab_size": 51865
 
46
  "scale_embedding": false,
47
  "suppress_tokens": [],
48
  "torch_dtype": "float32",
49
+ "transformers_version": "4.37.0.dev0",
50
  "use_cache": true,
51
  "use_weighted_layer_sum": false,
52
  "vocab_size": 51865
generation_config.json CHANGED
@@ -243,5 +243,5 @@
243
  "transcribe": 50359,
244
  "translate": 50358
245
  },
246
- "transformers_version": "4.36.2"
247
  }
 
243
  "transcribe": 50359,
244
  "translate": 50358
245
  },
246
+ "transformers_version": "4.37.0.dev0"
247
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6e6e0de0a3f53660356543fe1d36d502c31d666b9dd16c3a2e7b559571da7a18
3
  size 3055544304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8deb0f819e0b09d469076a8b146d2306b600fb99a18bb942919400303f6526a
3
  size 3055544304
runs/Jan29_21-18-20_DESKTOP-N7AL96B/events.out.tfevents.1706555902.DESKTOP-N7AL96B ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ebbe360ad1da1d5e62b59f21cf82837556580c92ade84806ddcc75581816dab
3
+ size 9225
runs/Jan29_22-54-50_DESKTOP-N7AL96B/events.out.tfevents.1706561691.DESKTOP-N7AL96B ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:454a9d0ea4c87ab0150f5f5ecf1047beae98f11f65ac1aeb8607b8384fe343e3
3
+ size 19872
runs/Jan30_20-30-34_DESKTOP-N7AL96B/events.out.tfevents.1706639435.DESKTOP-N7AL96B ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c322238ca6f6e1aa867656256bf22d6cd9258062e8e59b5549840ee584ed0c8b
3
+ size 19872
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ef139039e7bc7d6c0fa2913e07d5591210bb5f1296d17addebefdaa5060429f5
3
- size 4475
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f93720c8b907e66d58e89a123da7cd40910bfd1ce3534b3747f2bb01c1f0d677
3
+ size 4920