Model save

README.md

@@ -5,9 +5,24 @@ tags:
 - generated_from_trainer
 datasets:
 - audiofolder
+metrics:
+- wer
 model-index:
 - name: whisper-medium-ft-custom
-  results: []
+  results:
+  - task:
+      name: Automatic Speech Recognition
+      type: automatic-speech-recognition
+    dataset:
+      name: audiofolder
+      type: audiofolder
+      config: default
+      split: None
+      args: default
+    metrics:
+    - name: Wer
+      type: wer
+      value: 100.0
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -16,6 +31,9 @@ should probably proofread and complete it, then remove this comment. -->
 # whisper-medium-ft-custom
 
 This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the audiofolder dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.4298
+- Wer: 100.0
 
 ## Model description
 
@@ -35,15 +53,25 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 1e-06
-- train_batch_size: 16
+- train_batch_size: 4
 - eval_batch_size: 8
 - seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 16
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
 - training_steps: 1000
 - mixed_precision_training: Native AMP
 
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Wer   |
+|:-------------:|:-----:|:----:|:---------------:|:-----:|
+| 0.0045        | 35.71 | 500  | 1.3081          | 100.0 |
+| 0.0012        | 71.43 | 1000 | 1.4298          | 100.0 |
+
+
 ### Framework versions
 
 - Transformers 4.37.2

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44ef6cf049375c53d2549c35327323f71c45ae7949638ccff1d77f8578570cc1
+oid sha256:f40521ac38a3efa0092ba457f80c72d089fd07a615e10d4896bdbe37f1fea60e
 size 3055544304