End of training

README.md

@@ -7,7 +7,7 @@ tags:
 - hf-asr-leaderboard
 - generated_from_trainer
 datasets:
-- common_voice_14_0
+- common_voice_16_1
 model-index:
 - name: ft_model
   results: []
@@ -20,8 +20,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the common_voice dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6208
-- Cer: 20.8328
+- Loss: 0.5946
+- Cer: 16.1162
 
 ## Model description
 
@@ -47,17 +47,15 @@ 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: 500
-- training_steps: 4000
+- training_steps: 2000
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Cer     |
 |:-------------:|:-----:|:----:|:---------------:|:-------:|
-| 0.5645        | 0.24  | 1000 | 0.7038          | 49.9136 |
-| 0.5042        | 0.48  | 2000 | 0.6748          | 30.8312 |
-| 0.5075        | 0.73  | 3000 | 0.6371          | 20.6798 |
-| 0.437         | 0.97  | 4000 | 0.6208          | 20.8328 |
+| 0.7715        | 0.05  | 1000 | 0.6407          | 19.5562 |
+| 0.4602        | 0.1   | 2000 | 0.5946          | 16.1162 |
 
 
 ### Framework versions