Hanhpt23
/

whisper-small-multimed

Automatic Speech Recognition

Generated from Trainer

Inference Endpoints

Model card Files Files and versions Community

Hanhpt23 commited on Jul 30

Commit

cde0c3b

•

1 Parent(s): c609d13

End of training

Files changed (1) hide show

README.md +7 -7

README.md CHANGED Viewed

@@ -17,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Hanhpt23/MultiMed dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7193
-- Wer: 19.9744
-- Cer: 14.0186
 ## Model description
@@ -51,10 +51,10 @@ The following hyperparameters were used during training:
 | Training Loss | Epoch | Step  | Validation Loss | Wer     | Cer     |
 |:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|
-| 0.6396        | 1.0   | 4626  | 0.7132          | 31.7615 | 22.7290 |
-| 0.3406        | 2.0   | 9252  | 0.6568          | 25.4179 | 18.0443 |
-| 0.1281        | 3.0   | 13878 | 0.6780          | 22.2623 | 15.5534 |
-| 0.026         | 4.0   | 18504 | 0.7193          | 19.9744 | 14.0186 |
 ### Framework versions

 This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Hanhpt23/MultiMed dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.7176
+- Wer: 20.2138
+- Cer: 14.0973
 ## Model description
 | Training Loss | Epoch | Step  | Validation Loss | Wer     | Cer     |
 |:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|
+| 0.5854        | 1.0   | 4626  | 0.7068          | 33.7558 | 24.6972 |
+| 0.3884        | 2.0   | 9252  | 0.6462          | 26.2113 | 19.0993 |
+| 0.119         | 3.0   | 13878 | 0.6747          | 21.5288 | 15.1996 |
+| 0.0365        | 4.0   | 18504 | 0.7176          | 20.2138 | 14.0973 |
 ### Framework versions