End of training

README.md

@@ -17,6 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 # demo_text_to_speech
 
 This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the audiofolder dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4246
 
 ## Model description
 
@@ -44,9 +46,21 @@ The following hyperparameters were used during training:
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
-- training_steps: 1000
+- training_steps: 1200
 - mixed_precision_training: Native AMP
 
+### Training results
+
+| Training Loss | Epoch   | Step | Validation Loss |
+|:-------------:|:-------:|:----:|:---------------:|
+| 0.5859        | 16.7111 | 200  | 0.5409          |
+| 0.479         | 33.3556 | 400  | 0.4511          |
+| 0.4312        | 50.0    | 600  | 0.4278          |
+| 0.4349        | 66.7111 | 800  | 0.4226          |
+| 0.4294        | 83.3556 | 1000 | 0.4225          |
+| 0.4146        | 100.0   | 1200 | 0.4246          |
+
+
 ### Framework versions
 
 - Transformers 4.50.3

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4b3426c93c295ff3ba0c768a8efb0c2ecaf50db5af30e56e3f3fda3709f5528b
+oid sha256:84473b1c7e9b33eb27137cb176e25a077297e51dad5868b24f681dfdac18016e
 size 577789320

runs/Apr07_14-46-30_332971cd1456/events.out.tfevents.1744037199.332971cd1456.214.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:57f1fda70544c916ce17cca0d35ec7eb1283bde1bc1de713574217c49462ee98
-size 18388
+oid sha256:ea2c0ab42ddede487a0c65c0b726785a94150ac56185cd1f0108be78e0a5ff67
+size 18742