Training finished

README.md

@@ -20,9 +20,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the quran-ayat-speech-to-text dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0092
-- Wer: 0.0825
-- Cer: 0.0329
+- Loss: 0.0071
+- Wer: 0.0856
+- Cer: 0.0333
 
 ## Model description
 
@@ -41,37 +41,31 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 5e-06
+- learning_rate: 0.0001
 - train_batch_size: 16
 - eval_batch_size: 16
 - seed: 42
 - gradient_accumulation_steps: 2
 - total_train_batch_size: 32
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
-- num_epochs: 10
+- num_epochs: 25
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Wer    | Cer    |
 |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
-| 0.0106        | 1.0   | 344  | 0.0116          | 0.0901 | 0.0326 |
-| 0.0083        | 2.0   | 688  | 0.0080          | 0.0916 | 0.0348 |
-| 0.0067        | 3.0   | 1032 | 0.0079          | 0.0880 | 0.0340 |
-| 0.004         | 4.0   | 1376 | 0.0080          | 0.0894 | 0.0344 |
-| 0.0033        | 5.0   | 1720 | 0.0084          | 0.0871 | 0.0326 |
-| 0.0034        | 6.0   | 2064 | 0.0087          | 0.0863 | 0.0326 |
-| 0.002         | 7.0   | 2408 | 0.0091          | 0.0865 | 0.0324 |
-| 0.0021        | 8.0   | 2752 | 0.0094          | 0.0852 | 0.0325 |
-| 0.0018        | 9.0   | 3096 | 0.0092          | 0.0825 | 0.0329 |
-| 0.0015        | 10.0  | 3440 | 0.0097          | 0.0843 | 0.0324 |
+| 0.0071        | 1.0   | 157  | 0.0073          | 0.0863 | 0.0323 |
+| 0.0044        | 2.0   | 314  | 0.0087          | 0.0992 | 0.0348 |
+| 0.0028        | 3.0   | 471  | 0.0102          | 0.1111 | 0.0399 |
+| 0.0038        | 4.0   | 628  | 0.0116          | 0.1209 | 0.0419 |
 
 
 ### Framework versions
 
-- Transformers 4.44.2
-- Pytorch 2.4.1+cu121
+- Transformers 4.47.0
+- Pytorch 2.5.1+cu121
 - Datasets 3.2.0
-- Tokenizers 0.19.1
+- Tokenizers 0.21.0

generation_config.json

@@ -244,5 +244,5 @@
     "transcribe": 50359,
     "translate": 50358
   },
-  "transformers_version": "4.44.2"
+  "transformers_version": "4.47.0"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71f7c9d3720fc46abecaa18a5658948d75504200297ba0c5114e0cda1624f1e1
+oid sha256:e39284a2070116bc4b8165f23c2d522585f1c0f76479bfe0a7e9a366406cfefa
 size 151061672

runs/Jan16_10-56-11_b646aa6c2161/events.out.tfevents.1737028396.b646aa6c2161.31.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd422f34cbeb4ddfda7941c91cdf44ea27d2ab3bdc9814efab1dbd4703671b33
+size 453