Upload README.md

README.md

@@ -1,6 +1,6 @@
 ---
 license: apache-2.0
-base_model: odunola/yoruba_whisper
+base_model: openai/whisper-small
 tags:
 - generated_from_trainer
 metrics:
@@ -15,10 +15,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # yoruba_whisper
 
-This model is a fine-tuned version of [odunola/yoruba_whisper](https://huggingface.co/odunola/yoruba_whisper) on an unknown dataset.
+This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4023
-- Wer: 49.2215
+- Loss: 0.5389
+- Wer: 56.0826
 
 ## Model description
 
@@ -44,15 +44,16 @@ 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: 2000
+- training_steps: 3000
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Wer     |
 |:-------------:|:-----:|:----:|:---------------:|:-------:|
-| 0.3182        | 1.85  | 1000 | 0.4067          | 50.5848 |
-| 0.1219        | 3.7   | 2000 | 0.4023          | 49.2215 |
+| 0.5342        | 1.85  | 1000 | 0.5837          | 60.3821 |
+| 0.2361        | 3.7   | 2000 | 0.5227          | 58.5009 |
+| 0.1185        | 5.56  | 3000 | 0.5389          | 56.0826 |
 
 
 ### Framework versions