End of training

README.md

@@ -1,40 +1,42 @@
 ---
+language:
+- dv
 license: apache-2.0
 base_model: openai/whisper-small
 tags:
 - generated_from_trainer
 datasets:
-- audiofolder
+- mozilla-foundation/common_voice_13_0
 metrics:
 - wer
 model-index:
-- name: test-model
+- name: custom_data_test
   results:
   - task:
       name: Automatic Speech Recognition
       type: automatic-speech-recognition
     dataset:
-      name: audiofolder
-      type: audiofolder
+      name: Common Voice 13
+      type: mozilla-foundation/common_voice_13_0
       config: custom_data
       split: test
       args: custom_data
     metrics:
     - name: Wer
       type: wer
-      value: 83.33333333333334
+      value: 8.333333333333332
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# test-model
+# custom_data_test
 
-This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the audiofolder dataset.
+This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 13 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.1527
-- Wer Ortho: 91.6667
-- Wer: 83.3333
+- Loss: 1.1513
+- Wer Ortho: 8.3333
+- Wer: 8.3333
 
 ## Model description
 
@@ -59,15 +61,14 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: constant_with_warmup
-- lr_scheduler_warmup_steps: 1
-- training_steps: 4
+- lr_scheduler_warmup_steps: 2
+- training_steps: 20
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer     |
-|:-------------:|:-----:|:----:|:---------------:|:---------:|:-------:|
-| 4.6983        | 0.29  | 2    | 3.6278          | 91.6667   | 83.3333 |
-| 3.2746        | 0.57  | 4    | 2.1527          | 91.6667   | 83.3333 |
+| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer    |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|
+| 1.1613        | 2.71  | 19   | 1.1513          | 8.3333    | 8.3333 |
 
 
 ### Framework versions

runs/Feb27_14-43-44_Js-MacBook-Pro.lan/events.out.tfevents.1709037829.Js-MacBook-Pro.lan.43788.2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad8abd9bc340265c15c32a62a7bb0a645ab3217b19bc19c232afc14b5447697f
-size 8560
+oid sha256:e6a13b38d2cde0d9c5274af1cd916b48fec5b55f1d8b40e9abc4df9556c5e574
+size 9115