End of training

README.md

@@ -20,9 +20,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [CKSINGH/whisper-small-hi-firefox](https://huggingface.co/CKSINGH/whisper-small-hi-firefox) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4751
-- Wer: 37.6174
-- Cer: 18.8637
+- Loss: 0.3893
+- Wer: 26.1932
+- Cer: 12.5576
 
 ## Model description
 
@@ -47,15 +47,17 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 1
-- training_steps: 1
+- lr_scheduler_warmup_steps: 500
+- training_steps: 3000
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Wer     | Cer     |
 |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
-| 0.4479        | 0.02  | 1    | 0.4751          | 37.6174 | 18.8637 |
+| 0.0115        | 17.24 | 1000 | 0.3435          | 26.9823 | 13.1975 |
+| 0.0004        | 34.48 | 2000 | 0.3805          | 26.4186 | 12.5888 |
+| 0.0001        | 51.72 | 3000 | 0.3893          | 26.1932 | 12.5576 |
 
 
 ### Framework versions