update model card README.md

README.md

@@ -14,8 +14,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4440
-- Wer: 0.3433
+- Loss: 0.4526
+- Wer: 0.3411
 
 ## Model description
 
@@ -48,13 +48,13 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Wer    |
 |:-------------:|:-----:|:----:|:---------------:|:------:|
-| 3.6205        | 4.0   | 500  | 1.4849          | 1.0057 |
-| 0.8238        | 8.0   | 1000 | 0.4860          | 0.4841 |
-| 0.297         | 12.0  | 1500 | 0.4335          | 0.4018 |
-| 0.17          | 16.0  | 2000 | 0.4707          | 0.3734 |
-| 0.1258        | 20.0  | 2500 | 0.4850          | 0.3618 |
-| 0.1063        | 24.0  | 3000 | 0.4798          | 0.3521 |
-| 0.0806        | 28.0  | 3500 | 0.4440          | 0.3433 |
+| 3.7503        | 4.0   | 500  | 2.4125          | 1.0006 |
+| 0.9595        | 8.0   | 1000 | 0.4833          | 0.4776 |
+| 0.3018        | 12.0  | 1500 | 0.4333          | 0.4062 |
+| 0.1751        | 16.0  | 2000 | 0.4474          | 0.3697 |
+| 0.1288        | 20.0  | 2500 | 0.4445          | 0.3558 |
+| 0.1073        | 24.0  | 3000 | 0.4695          | 0.3464 |
+| 0.0816        | 28.0  | 3500 | 0.4526          | 0.3411 |
 
 
 ### Framework versions