Update README.md

README.md

@@ -1,4 +1,9 @@
 ---
+language:
+- cs
+tags:
+- audio
+- speech
 dataset_info:
   features:
   - name: file_name
@@ -7,6 +12,22 @@ dataset_info:
     dtype: string
   - name: audio
     dtype: audio
-language:
-- cs
----
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: train.parquet
+---
+
+# Czech Audio Dataset
+
+This dataset contains audio files and transcriptions.
+
+- **Total Size:** ~6.3 GB
+- **Language:** Czech (cs)
+- **Format:** Parquet (Audio bytes embedded)
+
+## Structure
+- `audio`: Contains the raw audio data (playable in viewer).
+- `transcription`: Text transcription of the audio.
+- `file_name`: Original filename.