Create README.md

README.md

@@ -0,0 +1,41 @@
+---
+base_model:
+- openai/whisper-small
+---
+Note: This classifier also contains fine-tuned `whisper-small` weights in its state dict. It will be properly loaded by my model wrapper.
+
+Result of the classifier Rob's human-annotated dataset (`data/voicemail_human_eval.csv`):
+
+```bash
+Results for chunk size 1 seconds:
+ - Accuracy: 0.7280
+ - Precision: 0.8435
+ - Recall: 0.7337
+ - F1 Score: 0.7848 
+ 
+
+Results for chunk size 2 seconds:
+ - Accuracy: 0.8120
+ - Precision: 0.9013
+ - Recall: 0.8107
+ - F1 Score: 0.8536 
+ 
+Results for chunk size 5 seconds:
+ - Accuracy: 0.8520
+ - Precision: 0.9459
+ - Recall: 0.8284
+ - F1 Score: 0.8833 
+ 
+Results for chunk size 10 seconds:
+ - Accuracy: 0.8840
+ - Precision: 0.9930
+ - Recall: 0.8343
+ - F1 Score: 0.9068 
+ 
+Results for full audio samples:
+ - Accuracy: 0.8760
+ - Precision: 0.9929
+ - Recall: 0.8225
+ - F1 Score: 0.8997
+```
+