Update README.md

This dataset combines the following two publicly available datasets for Korean language:

https://huggingface.co/datasets/Junhoee/STT_Korean_Dataset_80000
and
https://huggingface.co/datasets/Bingsu/zeroth-korean

Combined, this offers over 100K rows of data for automatic speech recognition training and fine-tuning.

README.md

@@ -1,22 +1,28 @@
----
-license: apache-2.0
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
-dataset_info:
-  features:
-  - name: audio
-    dtype:
-      audio:
-        sampling_rate: 16000
-  - name: text
-    dtype: string
-  splits:
-  - name: train
-    num_bytes: 11169873357.152
-    num_examples: 102263
-  download_size: 10186919269
-  dataset_size: 11169873357.152
----
+---
+license: apache-2.0
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
+dataset_info:
+  features:
+  - name: audio
+    dtype:
+      audio:
+        sampling_rate: 16000
+  - name: text
+    dtype: string
+  splits:
+  - name: train
+    num_bytes: 11169873357.152
+    num_examples: 102263
+  download_size: 10186919269
+  dataset_size: 11169873357.152
+task_categories:
+- automatic-speech-recognition
+language:
+- ko
+size_categories:
+- 100K<n<1M
+---