deepdml commited on
Commit
9d834e8
·
verified ·
1 Parent(s): 35c0ec8

End of training

Browse files
Files changed (2) hide show
  1. README.md +10 -22
  2. model.safetensors +1 -1
README.md CHANGED
@@ -26,8 +26,9 @@ model-index:
26
  metrics:
27
  - name: Wer
28
  type: wer
29
- value: 34.648462173849666
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
 
@@ -35,9 +36,9 @@ should probably proofread and complete it, then remove this comment. -->
35
 
36
  This model is a fine-tuned version of [deepdml/whisper-large-v3-turbo](https://huggingface.co/deepdml/whisper-large-v3-turbo) on the google/fleurs dataset.
37
  It achieves the following results on the evaluation set:
38
- - Loss: 0.7047
39
- - Wer: 34.6485
40
- - Cer: 10.6743
41
 
42
  ## Model description
43
 
@@ -69,11 +70,11 @@ The following hyperparameters were used during training:
69
 
70
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
71
  |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
72
- | 0.2017 | 0.2 | 1000 | 0.6332 | 40.5520 | 13.2337 |
73
- | 0.1311 | 0.4 | 2000 | 0.6560 | 37.9865 | 12.0943 |
74
- | 0.0572 | 0.6 | 3000 | 0.6759 | 36.4608 | 11.5819 |
75
- | 0.0503 | 0.8 | 4000 | 0.6915 | 35.1684 | 10.8913 |
76
- | 0.0355 | 1.0 | 5000 | 0.7047 | 34.6485 | 10.6743 |
77
 
78
 
79
  ### Framework versions
@@ -82,16 +83,3 @@ The following hyperparameters were used during training:
82
  - Pytorch 2.3.0+cu121
83
  - Datasets 2.19.1
84
  - Tokenizers 0.19.1
85
-
86
- ## Citation
87
-
88
- Please cite the model using the following BibTeX entry:
89
-
90
- ```bibtex
91
- @misc{deepdml/whisper-large-v3-turbo-ig-mix-norm,
92
- title={Fine-tuned Whisper turbo ASR model for speech recognition in Lingala},
93
- author={Jimenez, David},
94
- howpublished={\url{https://huggingface.co/deepdml/whisper-large-v3-turbo-ig-mix-norm}},
95
- year={2025}
96
- }
97
- ```
 
26
  metrics:
27
  - name: Wer
28
  type: wer
29
+ value: 31.264605428725506
30
  ---
31
+
32
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
33
  should probably proofread and complete it, then remove this comment. -->
34
 
 
36
 
37
  This model is a fine-tuned version of [deepdml/whisper-large-v3-turbo](https://huggingface.co/deepdml/whisper-large-v3-turbo) on the google/fleurs dataset.
38
  It achieves the following results on the evaluation set:
39
+ - Loss: 0.7028
40
+ - Wer: 31.2646
41
+ - Cer: 10.8084
42
 
43
  ## Model description
44
 
 
70
 
71
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
72
  |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
73
+ | 0.2019 | 0.2 | 1000 | 0.6438 | 36.4596 | 12.5223 |
74
+ | 0.1293 | 0.4 | 2000 | 0.6558 | 33.7633 | 11.6044 |
75
+ | 0.0589 | 0.6 | 3000 | 0.6882 | 31.8758 | 10.6653 |
76
+ | 0.0504 | 0.8 | 4000 | 0.6845 | 31.0669 | 10.3172 |
77
+ | 0.0353 | 1.0 | 5000 | 0.7028 | 31.2646 | 10.8084 |
78
 
79
 
80
  ### Framework versions
 
83
  - Pytorch 2.3.0+cu121
84
  - Datasets 2.19.1
85
  - Tokenizers 0.19.1
 
 
 
 
 
 
 
 
 
 
 
 
 
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fa2afba5225d169b03c2904cd045e593b67af384cb6bfb87115e5ee3996c5f4a
3
  size 3235581408
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b31b8a3d88dc0333219d41774509c5550880704a7016acda787a7ff238e1318
3
  size 3235581408