End of training

README.md

@@ -14,16 +14,16 @@ model-index:
 - name: Whisper Small Bashkir
   results:
   - task:
-      type: automatic-speech-recognition
       name: Automatic Speech Recognition
+      type: automatic-speech-recognition
     dataset:
       name: Common Voice 17.0 (ba)
       type: stdbug/common-voice-17-ba
       args: 'config: ba, split: test'
     metrics:
-    - type: wer
-      value: 24.803682642837803
-      name: Wer
+    - name: Wer
+      type: wer
+      value: 24.55759888965996
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,8 +33,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 17.0 (ba) dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1652
-- Wer: 24.8037
+- Loss: 0.1616
+- Wer: 24.5576
 
 ## Model description
 
@@ -60,14 +60,14 @@ The following hyperparameters were used during training:
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
-- training_steps: 13842
+- training_steps: 19576
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch  | Step  | Validation Loss | Wer     |
 |:-------------:|:------:|:-----:|:---------------:|:-------:|
-| 0.1363        | 0.9999 | 13842 | 0.1652          | 24.8037 |
+| 0.0861        | 0.9999 | 19576 | 0.1616          | 24.5576 |
 
 
 ### Framework versions

generation_config.json

@@ -150,7 +150,7 @@
     "<|yo|>": 50325,
     "<|zh|>": 50260
   },
-  "language": "bashkir",
+  "language": "Bashkir",
   "max_initial_timestamp_index": 50,
   "max_length": 448,
   "no_timestamps_token_id": 50363,