Hugging Face's logo

makhataei
/
Whisper-Small-Common-Voice

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Persian
whisper
fa-asr
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
makhataei commited on
Commit
a6d87bb
•
1 Parent(s): ab34a02

End of training

Browse files
Files changed (3) hide show
  1. README.md +33 -23
  2. model.safetensors +1 -1
  3. runs/Jan03_17-20-31_Software-AI/events.out.tfevents.1704289834.Software-AI.2159775.0 +2 -2
README.md CHANGED
@@ -22,8 +22,8 @@ should probably proofread and complete it, then remove this comment. -->
22
 
23
  This model is a fine-tuned version of [makhataei/Whisper-Small-Common-Voice](https://huggingface.co/makhataei/Whisper-Small-Common-Voice) on the Common Voice 16.0 dataset.
24
  It achieves the following results on the evaluation set:
25
- - Loss: 0.7543
26
- - Wer: 46.0283
27
 
28
  ## Model description
29
 
@@ -51,32 +51,42 @@ The following hyperparameters were used during training:
51
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
  - lr_scheduler_type: linear
53
  - lr_scheduler_warmup_steps: 50
54
- - training_steps: 2000
55
 
56
  ### Training results
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Wer |
59
  |:-------------:|:-----:|:----:|:---------------:|:-------:|
60
- | 0.0137 | 0.16 | 100 | 0.7336 | 48.8685 |
61
- | 0.0153 | 0.31 | 200 | 0.7163 | 45.0962 |
62
- | 0.016 | 0.47 | 300 | 0.7235 | 45.9726 |
63
- | 0.0133 | 0.62 | 400 | 0.7356 | 45.7130 |
64
- | 0.0186 | 0.78 | 500 | 0.6912 | 44.0760 |
65
- | 0.012 | 0.93 | 600 | 0.6939 | 44.9664 |
66
- | 0.0092 | 1.09 | 700 | 0.7184 | 44.8180 |
67
- | 0.0071 | 1.25 | 800 | 0.7321 | 44.0691 |
68
- | 0.0072 | 1.4 | 900 | 0.7485 | 46.3714 |
69
- | 0.0061 | 1.56 | 1000 | 0.7180 | 44.9502 |
70
- | 0.0063 | 1.71 | 1100 | 0.7036 | 46.2741 |
71
- | 0.004 | 1.87 | 1200 | 0.7296 | 45.0684 |
72
- | 0.0036 | 2.03 | 1300 | 0.7275 | 46.3158 |
73
- | 0.002 | 2.18 | 1400 | 0.7432 | 45.9332 |
74
- | 0.0014 | 2.34 | 1500 | 0.7472 | 45.5692 |
75
- | 0.0012 | 2.49 | 1600 | 0.7408 | 44.7902 |
76
- | 0.0201 | 2.65 | 1700 | 0.7566 | 45.7593 |
77
- | 0.0011 | 2.8 | 1800 | 0.7551 | 45.3327 |
78
- | 0.0011 | 2.96 | 1900 | 0.7550 | 46.1720 |
79
- | 0.0004 | 3.12 | 2000 | 0.7543 | 46.0283 |
 
 
 
 
 
 
 
 
 
 
80
 
81
 
82
  ### Framework versions
 
22
 
23
  This model is a fine-tuned version of [makhataei/Whisper-Small-Common-Voice](https://huggingface.co/makhataei/Whisper-Small-Common-Voice) on the Common Voice 16.0 dataset.
24
  It achieves the following results on the evaluation set:
25
+ - Loss: 0.7773
26
+ - Wer: 58.2147
27
 
28
  ## Model description
29
 
 
51
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
  - lr_scheduler_type: linear
53
  - lr_scheduler_warmup_steps: 50
54
+ - training_steps: 3000
55
 
56
  ### Training results
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Wer |
59
  |:-------------:|:-----:|:----:|:---------------:|:-------:|
60
+ | 0.0106 | 0.16 | 100 | 0.7024 | 44.3125 |
61
+ | 0.0107 | 0.31 | 200 | 0.7246 | 46.4665 |
62
+ | 0.0121 | 0.47 | 300 | 0.7213 | 44.8249 |
63
+ | 0.0096 | 0.62 | 400 | 0.7252 | 48.7294 |
64
+ | 0.0118 | 0.78 | 500 | 0.7118 | 47.6837 |
65
+ | 0.011 | 0.93 | 600 | 0.7221 | 47.6119 |
66
+ | 0.0058 | 1.09 | 700 | 0.7494 | 46.9117 |
67
+ | 0.0051 | 1.25 | 800 | 0.7611 | 46.6844 |
68
+ | 0.009 | 1.4 | 900 | 0.7271 | 47.2919 |
69
+ | 0.0077 | 1.56 | 1000 | 0.7287 | 58.6274 |
70
+ | 0.0058 | 1.71 | 1100 | 0.7582 | 64.5653 |
71
+ | 0.0067 | 1.87 | 1200 | 0.7135 | 66.7841 |
72
+ | 0.0031 | 2.03 | 1300 | 0.7395 | 61.1755 |
73
+ | 0.0033 | 2.18 | 1400 | 0.7353 | 62.7568 |
74
+ | 0.0025 | 2.34 | 1500 | 0.7535 | 68.1080 |
75
+ | 0.0022 | 2.49 | 1600 | 0.7521 | 63.5126 |
76
+ | 0.0151 | 2.65 | 1700 | 0.7546 | 71.5140 |
77
+ | 0.0022 | 2.8 | 1800 | 0.7470 | 79.6058 |
78
+ | 0.0017 | 2.96 | 1900 | 0.7286 | 70.0348 |
79
+ | 0.0008 | 3.12 | 2000 | 0.7544 | 66.4433 |
80
+ | 0.0007 | 3.27 | 2100 | 0.7618 | 60.4336 |
81
+ | 0.0004 | 3.43 | 2200 | 0.7672 | 64.8505 |
82
+ | 0.0015 | 3.58 | 2300 | 0.7853 | 60.2017 |
83
+ | 0.0008 | 3.74 | 2400 | 0.7719 | 59.2233 |
84
+ | 0.0003 | 3.9 | 2500 | 0.7537 | 60.1414 |
85
+ | 0.0002 | 4.05 | 2600 | 0.7656 | 59.9490 |
86
+ | 0.0087 | 4.21 | 2700 | 0.7726 | 59.2140 |
87
+ | 0.0003 | 4.36 | 2800 | 0.7750 | 59.6661 |
88
+ | 0.0002 | 4.52 | 2900 | 0.7766 | 58.8036 |
89
+ | 0.0002 | 4.67 | 3000 | 0.7773 | 58.2147 |
90
 
91
 
92
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2a6ef70d788679f8f42437e3fa266ec0cee39538166fc48ae86e5191610b7e2e
3
  size 966995080
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41800be11466937a50cf34d4eb1b014f0f9539d89902cba3bafa6672a6c8bd3d
3
  size 966995080
runs/Jan03_17-20-31_Software-AI/events.out.tfevents.1704289834.Software-AI.2159775.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7e0ff8f2548445143c5756b687695ea5810373010665cc46938529edbc2393d4
3
- size 32495
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:25ce6846b88e55de7283dba1b803ffe4cf66dd8424fccb197560223d7228918c
3
+ size 33795