End of training

README.md

@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the facebook/voxpopuli dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6011
+- Loss: 0.4581
 
 ## Model description
 
@@ -44,24 +44,18 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 32
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 50
-- training_steps: 100
+- lr_scheduler_warmup_steps: 500
+- training_steps: 4000
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss |
-|:-------------:|:------:|:----:|:---------------:|
-| No log        | 0.0430 | 10   | 0.7645          |
-| No log        | 0.0861 | 20   | 0.7302          |
-| 0.8445        | 0.1291 | 30   | 0.6879          |
-| 0.8445        | 0.1721 | 40   | 0.6481          |
-| 0.7324        | 0.2152 | 50   | 0.6331          |
-| 0.7324        | 0.2582 | 60   | 0.6189          |
-| 0.7324        | 0.3012 | 70   | 0.6095          |
-| 0.6788        | 0.3443 | 80   | 0.6067          |
-| 0.6788        | 0.3873 | 90   | 0.6032          |
-| 0.679         | 0.4303 | 100  | 0.6011          |
+| Training Loss | Epoch   | Step | Validation Loss |
+|:-------------:|:-------:|:----:|:---------------:|
+| 0.5201        | 4.3034  | 1000 | 0.4770          |
+| 0.4962        | 8.6068  | 2000 | 0.4638          |
+| 0.4926        | 12.9102 | 3000 | 0.4597          |
+| 0.4916        | 17.2136 | 4000 | 0.4581          |
 
 
 ### Framework versions