Training in progress, step 1000

README.md

@@ -25,17 +25,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on the JacobLinCool/common_voice_19_0_zh-TW dataset.
 It achieves the following results on the evaluation set:
-- eval_loss: 0.1246
-- eval_wer: 38.7592
-- eval_cer: 11.5327
-- eval_decode_runtime: 89.6947
-- eval_wer_runtime: 0.1256
-- eval_cer_runtime: 0.1558
-- eval_runtime: 332.3335
-- eval_samples_per_second: 15.084
+- eval_loss: 0.1210
+- eval_wer: 40.4548
+- eval_cer: 14.4620
+- eval_decode_runtime: 90.3624
+- eval_wer_runtime: 0.1421
+- eval_cer_runtime: 0.1701
+- eval_runtime: 332.9706
+- eval_samples_per_second: 15.055
 - eval_steps_per_second: 0.472
-- epoch: 0.1
-- step: 500
+- epoch: 0.2
+- step: 1000
 
 ## Model description
 

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4c0c33e6f71cd0b07393bd7f5af3756fb87cd10e3981cde6b44da52ab084fcb8
+oid sha256:9bfae3cc11c0a63139a0e88b8bea77b73219112554e5c15d8c86c535eb734669
 size 230831688

runs/Oct31_16-15-33_24a46c0bfe1b/events.out.tfevents.1730391672.24a46c0bfe1b.67146.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:50e835c9524043ed796366b7b80275b6f6eb7d74f0ec0cd230063212ae51d0e6
-size 641148
+oid sha256:f0c634f6ffe65c2eafa8855331e4bed4c2c03d471e65c0308b141ba2efc750f6
+size 990577