Training in progress, step 3300

README.md

@@ -16,13 +16,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the None dataset.
 It achieves the following results on the evaluation set:
-- eval_loss: 0.3947
-- eval_wer: 71.7241
-- eval_runtime: 145.8938
-- eval_samples_per_second: 0.685
-- eval_steps_per_second: 0.021
-- epoch: 8.9316
-- step: 3000
+- eval_loss: 0.4246
+- eval_wer: 72.5862
+- eval_runtime: 146.943
+- eval_samples_per_second: 0.681
+- eval_steps_per_second: 0.02
+- epoch: 9.9088
+- step: 3300
 
 ## Model description