Update README.md

README.md

@@ -17,9 +17,9 @@ This dataset contains audio recordings in Uzbek and their corresponding transcri
 The dataset is structured as follows:
 
 - **audio/**: Contains all audio files.
-  - **uz_other_dataset/**: Contains `.tar` files, each containing multiple `.wav` files.
-  - **uz_train_dataset/**: Contains `.tar` files, each containing multiple `.wav` files.
-  - **uz_validated_dataset/**: Contains `.tar` files, each containing multiple `.wav` files.
+  - **uz_other_dataset/**: Contains `.tar` files, each containing multiple `.mp3` files.
+  - **uz_train_dataset/**: Contains `.tar` files, each containing multiple `.mp3` files.
+  - **uz_validated_dataset/**: Contains `.tar` files, each containing multiple `.mp3` files.
   - **wavs/**: Contains individual `.wav` files.
 
 - **transcription/**: Contains `.tsv` files with transcriptions.