update model card README.md

README.md

@@ -4,7 +4,7 @@ base_model: openai/whisper-small
 tags:
 - generated_from_trainer
 datasets:
-- yt
+- common_voice_9_0
 metrics:
 - wer
 model-index:
@@ -14,13 +14,15 @@ model-index:
       name: Automatic Speech Recognition
       type: automatic-speech-recognition
     dataset:
-      name: yt id
-      type: yt
+      name: common_voice_9_0
+      type: common_voice_9_0
+      config: id
+      split: train
       args: id
     metrics:
     - name: Wer
       type: wer
-      value: 100.0
+      value: 5.376528641922334
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -28,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # yt-special-batch8
 
-This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the yt id dataset.
+This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the common_voice_9_0 dataset.
 It achieves the following results on the evaluation set:
-- Loss: nan
-- Wer: 100.0
+- Loss: 0.2578
+- Wer: 5.3765
 
 ## Model description
 
@@ -52,7 +54,7 @@ More information needed
 The following hyperparameters were used during training:
 - learning_rate: 1e-05
 - train_batch_size: 8
-- eval_batch_size: 8
+- eval_batch_size: 4
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
@@ -61,13 +63,13 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Wer   |
-|:-------------:|:-----:|:----:|:---------------:|:-----:|
-| 0.0           | 0.17  | 1000 | nan             | 100.0 |
-| 0.0           | 0.34  | 2000 | nan             | 100.0 |
-| 0.0           | 0.52  | 3000 | nan             | 100.0 |
-| 0.0           | 0.69  | 4000 | nan             | 100.0 |
-| 0.0           | 0.86  | 5000 | nan             | 100.0 |
+| Training Loss | Epoch | Step | Validation Loss | Wer      |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 37.1656       | 1.58  | 1000 | 31.3287         | 568.0305 |
+| 15.0344       | 3.17  | 2000 | 13.1680         | 145.2285 |
+| 7.6075        | 4.75  | 3000 | 5.8663          | 42.3536  |
+| 2.5225        | 6.34  | 4000 | 2.0001          | 19.5409  |
+| 0.5364        | 7.92  | 5000 | 0.2578          | 5.3765   |
 
 
 ### Framework versions