update model card README.md

README.md

@@ -1,9 +1,5 @@
 ---
-language:
-- fa
 tags:
-- automatic-speech-recognition
-- mozilla-foundation/common_voice_7_0
 - generated_from_trainer
 datasets:
 - common_voice
@@ -17,9 +13,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # common7
 
-This model is a fine-tuned version of [common7/checkpoint-4000](https://huggingface.co/common7/checkpoint-4000) on the MOZILLA-FOUNDATION/COMMON_VOICE_7_0 - FA dataset.
+This model was trained from scratch on the common_voice dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3428
+- Loss: 0.3441
 - Wer: 0.3493
 
 ## Model description
@@ -48,7 +44,7 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 100
-- num_epochs: 123.0
+- num_epochs: 150.0
 - mixed_precision_training: Native AMP
 
 ### Training results
@@ -92,6 +88,14 @@ The following hyperparameters were used during training:
 | 0.8667        | 115.13 | 17500 | 0.3438          | 0.3506 |
 | 0.8895        | 118.42 | 18000 | 0.3434          | 0.3503 |
 | 0.8888        | 121.71 | 18500 | 0.3425          | 0.3494 |
+| 0.9453        | 125.0  | 19000 | 0.3415          | 0.3480 |
+| 0.9267        | 128.29 | 19500 | 0.3477          | 0.3503 |
+| 0.8315        | 131.58 | 20000 | 0.3476          | 0.3505 |
+| 0.8542        | 134.86 | 20500 | 0.3475          | 0.3506 |
+| 0.8478        | 138.16 | 21000 | 0.3430          | 0.3481 |
+| 0.8643        | 141.45 | 21500 | 0.3451          | 0.3485 |
+| 0.8705        | 144.73 | 22000 | 0.3444          | 0.3474 |
+| 0.9869        | 148.03 | 22500 | 0.3441          | 0.3493 |
 
 
 ### Framework versions