End of training
Browse files- README.md +79 -4
- model.safetensors +1 -1
README.md
CHANGED
@@ -1,19 +1,48 @@
|
|
1 |
---
|
|
|
|
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: openai/whisper-medium
|
4 |
tags:
|
5 |
- generated_from_trainer
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6 |
model-index:
|
7 |
-
- name:
|
8 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
---
|
10 |
|
11 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
12 |
should probably proofread and complete it, then remove this comment. -->
|
13 |
|
14 |
-
#
|
15 |
|
16 |
-
This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on
|
|
|
|
|
|
|
|
|
|
|
17 |
|
18 |
## Model description
|
19 |
|
@@ -42,6 +71,52 @@ The following hyperparameters were used during training:
|
|
42 |
- training_steps: 4000
|
43 |
- mixed_precision_training: Native AMP
|
44 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
45 |
### Framework versions
|
46 |
|
47 |
- Transformers 4.41.2
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- ga
|
4 |
+
- en
|
5 |
license: apache-2.0
|
6 |
base_model: openai/whisper-medium
|
7 |
tags:
|
8 |
- generated_from_trainer
|
9 |
+
datasets:
|
10 |
+
- ymoslem/IWSLT2023-GA-EN
|
11 |
+
- ymoslem/FLEURS-GA-EN
|
12 |
+
- ymoslem/BitesizeIrish-GA-EN
|
13 |
+
- ymoslem/SpokenWords-GA-EN-MTed
|
14 |
+
metrics:
|
15 |
+
- bleu
|
16 |
+
- wer
|
17 |
model-index:
|
18 |
+
- name: Whisper Medium GA-EN Speech Translation Raw
|
19 |
+
results:
|
20 |
+
- task:
|
21 |
+
name: Automatic Speech Recognition
|
22 |
+
type: automatic-speech-recognition
|
23 |
+
dataset:
|
24 |
+
name: IWSLT-2023, FLEURS, BiteSize, and SpokenWords
|
25 |
+
type: ymoslem/IWSLT2023-GA-EN
|
26 |
+
metrics:
|
27 |
+
- name: Bleu
|
28 |
+
type: bleu
|
29 |
+
value: 26.56
|
30 |
+
- name: Wer
|
31 |
+
type: wer
|
32 |
+
value: 76.67717244484467
|
33 |
---
|
34 |
|
35 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
36 |
should probably proofread and complete it, then remove this comment. -->
|
37 |
|
38 |
+
# Whisper Medium GA-EN Speech Translation Raw
|
39 |
|
40 |
+
This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the IWSLT-2023, FLEURS, BiteSize, and SpokenWords dataset.
|
41 |
+
It achieves the following results on the evaluation set:
|
42 |
+
- Loss: 1.5187
|
43 |
+
- Bleu: 26.56
|
44 |
+
- Chrf: 46.91
|
45 |
+
- Wer: 76.6772
|
46 |
|
47 |
## Model description
|
48 |
|
|
|
71 |
- training_steps: 4000
|
72 |
- mixed_precision_training: Native AMP
|
73 |
|
74 |
+
### Training results
|
75 |
+
|
76 |
+
| Training Loss | Epoch | Step | Bleu | Chrf | Validation Loss | Wer |
|
77 |
+
|:-------------:|:------:|:----:|:-----:|:-----:|:---------------:|:--------:|
|
78 |
+
| 2.5874 | 0.0539 | 100 | 4.9 | 19.49 | 2.1785 | 114.0027 |
|
79 |
+
| 2.3237 | 0.1079 | 200 | 6.48 | 22.77 | 2.1129 | 151.8235 |
|
80 |
+
| 2.192 | 0.1618 | 300 | 7.92 | 25.9 | 2.0182 | 148.6718 |
|
81 |
+
| 1.9861 | 0.2157 | 400 | 10.55 | 28.55 | 1.8607 | 121.0266 |
|
82 |
+
| 1.8893 | 0.2697 | 500 | 16.68 | 33.64 | 1.8560 | 89.7794 |
|
83 |
+
| 1.8526 | 0.3236 | 600 | 8.83 | 30.12 | 1.7738 | 166.9968 |
|
84 |
+
| 1.6537 | 0.3776 | 700 | 10.94 | 33.83 | 1.6781 | 152.2287 |
|
85 |
+
| 1.7103 | 0.4315 | 800 | 16.9 | 36.4 | 1.6389 | 92.2557 |
|
86 |
+
| 1.4837 | 0.4854 | 900 | 13.81 | 34.5 | 1.6077 | 124.2233 |
|
87 |
+
| 1.2784 | 0.5394 | 1000 | 14.79 | 37.53 | 1.6103 | 116.3440 |
|
88 |
+
| 1.111 | 0.5933 | 1100 | 19.31 | 39.0 | 1.5579 | 93.6965 |
|
89 |
+
| 1.167 | 0.6472 | 1200 | 20.88 | 41.7 | 1.5210 | 91.6704 |
|
90 |
+
| 1.2217 | 0.7012 | 1300 | 21.29 | 41.72 | 1.4719 | 84.9167 |
|
91 |
+
| 1.0613 | 0.7551 | 1400 | 28.3 | 44.37 | 1.4663 | 67.1319 |
|
92 |
+
| 0.9256 | 0.8091 | 1500 | 27.5 | 45.59 | 1.4258 | 68.7078 |
|
93 |
+
| 0.8023 | 0.8630 | 1600 | 27.1 | 46.27 | 1.4027 | 72.7600 |
|
94 |
+
| 0.8327 | 0.9169 | 1700 | 27.03 | 46.19 | 1.3784 | 73.0302 |
|
95 |
+
| 0.7019 | 0.9709 | 1800 | 28.91 | 46.34 | 1.4127 | 67.4921 |
|
96 |
+
| 0.2681 | 1.0248 | 1900 | 28.53 | 47.12 | 1.3955 | 68.3026 |
|
97 |
+
| 0.2659 | 1.0787 | 2000 | 28.37 | 45.85 | 1.4194 | 68.1225 |
|
98 |
+
| 0.4202 | 1.1327 | 2100 | 1.5449| 27.53 | 44.0 | 69.8784 |
|
99 |
+
| 0.4212 | 1.1866 | 2200 | 1.6060| 25.89 | 43.05 | 70.1036 |
|
100 |
+
| 0.4124 | 1.2406 | 2300 | 1.6167| 24.31 | 41.55 | 75.8217 |
|
101 |
+
| 0.4696 | 1.2945 | 2400 | 1.5904| 21.79 | 41.86 | 85.0968 |
|
102 |
+
| 0.4018 | 1.3484 | 2500 | 1.6300| 25.36 | 43.45 | 76.4070 |
|
103 |
+
| 0.4549 | 1.4024 | 2600 | 1.5540| 26.06 | 44.27 | 71.9946 |
|
104 |
+
| 0.4018 | 1.4563 | 2700 | 1.5721| 26.22 | 45.42 | 72.9851 |
|
105 |
+
| 0.3534 | 1.5102 | 2800 | 1.5488| 23.65 | 44.43 | 80.0090 |
|
106 |
+
| 0.2907 | 1.5642 | 2900 | 1.5494| 24.04 | 42.57 | 75.3715 |
|
107 |
+
| 0.3117 | 1.6181 | 3000 | 1.5691| 28.27 | 45.06 | 67.2670 |
|
108 |
+
| 0.3379 | 1.6721 | 3100 | 1.4951| 30.52 | 47.42 | 65.5561 |
|
109 |
+
| 0.3686 | 1.7260 | 3200 | 1.5010| 30.7 | 48.13 | 64.8357 |
|
110 |
+
| 0.2855 | 1.7799 | 3300 | 1.5197| 27.19 | 46.47 | 74.5610 |
|
111 |
+
| 0.2919 | 1.8339 | 3400 | 1.4974| 31.39 | 48.56 | 63.5299 |
|
112 |
+
| 0.2582 | 1.8878 | 3500 | 1.4779| 30.18 | 48.54 | 64.9257 |
|
113 |
+
| 0.2523 | 1.9417 | 3600 | 1.4835| 30.29 | 47.07 | 66.6367 |
|
114 |
+
| 0.2005 | 1.9957 | 3700 | 1.4682| 29.89 | 47.95 | 68.2125 |
|
115 |
+
| 0.0617 | 2.0496 | 3800 | 1.5221| 29.49 | 47.1 | 67.6272 |
|
116 |
+
| 0.0661 | 2.1036 | 3900 | 1.5142| 26.93 | 46.91 | 75.8217 |
|
117 |
+
| 0.0609 | 2.1575 | 4000 | 1.5187| 26.56 | 46.91 | 76.6772 |
|
118 |
+
|
119 |
+
|
120 |
### Framework versions
|
121 |
|
122 |
- Transformers 4.41.2
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 3055544304
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c6527d1bad92936baac29a7501614271566650c6881c33c36ff6255dff805262
|
3 |
size 3055544304
|