./test-whisper-tiny-th

README.md

@@ -16,8 +16,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.8860
-- Cer: 35.4839
+- Loss: 0.8297
+- Cer: 33.9718
 
 ## Model description
 
@@ -48,11 +48,11 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Cer     |
 |:-------------:|:-----:|:----:|:---------------:|:-------:|
-| No log        | 1.0   | 7    | 0.9718          | 37.0968 |
-| 1.1407        | 2.0   | 14   | 0.9296          | 55.2419 |
-| 0.895         | 3.0   | 21   | 0.9023          | 53.9315 |
-| 0.895         | 4.0   | 28   | 0.8905          | 33.4677 |
-| 0.8158        | 5.0   | 35   | 0.8860          | 35.4839 |
+| No log        | 1.0   | 7    | 0.8606          | 33.9718 |
+| 0.6839        | 2.0   | 14   | 0.8408          | 34.4758 |
+| 0.5734        | 3.0   | 21   | 0.8370          | 33.2661 |
+| 0.5734        | 4.0   | 28   | 0.8336          | 34.1734 |
+| 0.5218        | 5.0   | 35   | 0.8297          | 33.9718 |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e9eb7edeaa5c7e7c3802a247eb057c7fc34ea17082f72db3baecb593ec2df0ed
+oid sha256:e0379ac062151cdd006e8ec32dfcf98fae4bc2dfd82fe67d49ec230be0b68df8
 size 151061672

runs/Sep11_07-08-20_b07ba3e2bb7a/events.out.tfevents.1726039654.b07ba3e2bb7a.204.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43eebcb7f1be07040da3adc32194261273d4d76fc5089e325dab0b189428df4e
+size 9227