End of training

README.md

@@ -1,6 +1,6 @@
 ---
 license: apache-2.0
-base_model: openai/whisper-medium
+base_model: PThi35/whisper_medium_phase2
 tags:
 - generated_from_trainer
 metrics:
@@ -15,11 +15,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 # testcode
 
-This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on an unknown dataset.
+This model is a fine-tuned version of [PThi35/whisper_medium_phase2](https://huggingface.co/PThi35/whisper_medium_phase2) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 4.0053
-- Cer: 18.4805
-- Wer: 34.2105
+- Loss: 0.8281
+- Cer: 22.7926
+- Wer: 36.8421
 
 ## Model description
 
@@ -54,8 +54,8 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss | Cer     | Wer     |
 |:-------------:|:------:|:----:|:---------------:|:-------:|:-------:|
-| 18.1291       | 0.6667 | 1    | 4.0053          | 18.4805 | 34.2105 |
-| 8.9428        | 2.0    | 3    | 4.0053          | 18.4805 | 34.2105 |
+| 2.027         | 0.6667 | 1    | 0.8281          | 22.7926 | 36.8421 |
+| 0.7125        | 2.0    | 3    | 0.8281          | 22.7926 | 36.8421 |
 
 
 ### Framework versions

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "openai/whisper-medium",
+  "_name_or_path": "PThi35/whisper_medium_phase2",
   "activation_dropout": 0.0,
   "activation_function": "gelu",
   "apply_spec_augment": true,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:62f73550fa6db24b0c6f6c5962bd0dae80fa644e93cde9cd9c3792971b47fd28
+oid sha256:97a84f989448b138eba9ec742ed31606edae5cf5d566283cc01295861f060cef
 size 3055544304

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:09f205f0ba163cf980efe1348b738ec1b1e9154d0c2ad859875d5884902aeaeb
+oid sha256:309b67248e46ba19658cb21d1f15457b17988e0ad02b7b677913238e44ac5831
 size 5240