faster-whisper-medium-nya

README.md

@@ -16,14 +16,14 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- eval_loss: 0.4360
-- eval_wer: 0.2618
-- eval_cer: 0.1012
-- eval_runtime: 589.3907
-- eval_samples_per_second: 1.055
-- eval_steps_per_second: 0.132
-- epoch: 1.4780
-- step: 3000
+- eval_loss: 0.3212
+- eval_wer: 0.2444
+- eval_cer: 0.0692
+- eval_runtime: 422.9731
+- eval_samples_per_second: 1.012
+- eval_steps_per_second: 0.128
+- epoch: 1.7130
+- step: 3477
 
 ## Model description
 

runs/Nov16_08-00-09_b54d33c26288/events.out.tfevents.1763280028.b54d33c26288.613.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0f5b801989327a8c046f916af22c543401d10847a5f01d9f6848eb0e40f454d1
-size 29443
+oid sha256:73262b053299b11b7a088678f5fedb0792eddeeaab1afdb6172729017a48960d
+size 29808