End of training

README.md

@@ -7,7 +7,7 @@ tags:
 - hf-asr-leaderboard
 - generated_from_trainer
 datasets:
-- ctam8736/papi_asr_mini
+- ctam8736/papi_asr
 metrics:
 - wer
 model-index:
@@ -18,12 +18,12 @@ model-index:
       type: automatic-speech-recognition
     dataset:
       name: Papi ASR Test
-      type: ctam8736/papi_asr_mini
+      type: ctam8736/papi_asr
       args: 'config: es, split: test'
     metrics:
     - name: Wer
       type: wer
-      value: 80.64516129032258
+      value: 22.26516425326875
 ---
 
 <!-- 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-tiny](https://huggingface.co/openai/whisper-tiny) on the Papi ASR Test dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.4408
-- Wer: 80.6452
+- Loss: 0.2038
+- Wer: 22.2652
 
 ## Model description
 
@@ -60,23 +60,18 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
-- training_steps: 100
+- training_steps: 1000
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Wer      |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|
-| No log        | 1.67  | 10   | 4.9708          | 94.8387  |
-| No log        | 3.33  | 20   | 4.8951          | 94.8387  |
-| 4.7977        | 5.0   | 30   | 4.6776          | 93.5484  |
-| 4.7977        | 6.67  | 40   | 4.3154          | 94.5161  |
-| 4.1417        | 8.33  | 50   | 3.9899          | 93.5484  |
-| 4.1417        | 10.0  | 60   | 3.7012          | 123.8710 |
-| 4.1417        | 11.67 | 70   | 3.4099          | 122.9032 |
-| 3.3662        | 13.33 | 80   | 3.0957          | 120.6452 |
-| 3.3662        | 15.0  | 90   | 2.7755          | 82.2581  |
-| 2.5332        | 16.67 | 100  | 2.4408          | 80.6452  |
+| Training Loss | Epoch | Step | Validation Loss | Wer     |
+|:-------------:|:-----:|:----:|:---------------:|:-------:|
+| 0.8909        | 0.34  | 200  | 0.8337          | 59.4875 |
+| 0.4264        | 0.69  | 400  | 0.4114          | 33.3977 |
+| 0.2452        | 1.03  | 600  | 0.2663          | 26.1676 |
+| 0.1953        | 1.38  | 800  | 0.2188          | 24.0369 |
+| 0.1699        | 1.72  | 1000 | 0.2038          | 22.2652 |
 
 
 ### Framework versions