Add dataset card with subset configurations

README.md

@@ -35,49 +35,11 @@ dataset_info:
         dtype:
           audio:
             sampling_rate: 16000
+      - name: is_denoised
+        dtype: int64
     splits:
     download_size: 0
     dataset_size: 0
-  - config_name: original
-    features:
-      - name: md5_text
-        dtype: string
-      - name: file_name
-        dtype: string
-      - name: md5_audio
-        dtype: string
-      - name: duration
-        dtype: float32
-      - name: text
-        dtype: string
-      - name: audio_language
-        dtype: string
-      - name: text_language
-        dtype: string
-      - name: session_id
-        dtype: string
-      - name: speaker
-        dtype: string
-      - name: type_concept
-        dtype: string
-      - name: recording_context
-        dtype: string
-      - name: medical_entities
-        dtype: string
-      - name: filename
-        dtype: string
-      - name: original_path
-        dtype: string
-      - name: audio
-        dtype:
-          audio:
-            sampling_rate: 16000
-    splits:
-      - name: test
-        num_bytes: 250000000
-        num_examples: 500
-    download_size: 212500000
-    dataset_size: 250000000
   - config_name: denoised
     features:
       - name: md5_text
@@ -112,6 +74,8 @@ dataset_info:
         dtype:
           audio:
             sampling_rate: 16000
+      - name: is_denoised
+        dtype: int64
     splits:
       - name: test
         num_bytes: 250000000
@@ -119,10 +83,6 @@ dataset_info:
     download_size: 212500000
     dataset_size: 250000000
 configs:
-  - config_name: original
-    data_files:
-      - split: test
-        path: original/test-*
   - config_name: denoised
     data_files:
       - split: test
@@ -137,16 +97,16 @@ tags:
   - audio
   - speech
   - asr
-pretty_name: Original Denoised Dataset
+pretty_name: Denoised Subset Fixed
 size_categories:
-  - 1K<n<10K
+  - n<1K
 ---
 
-# Original Denoised Dataset
+# Denoised Subset Fixed
 
 ## Dataset Description
 
-This dataset contains 1,000 samples organized across multiple splits and 2 subsets.
+This dataset contains 500 samples organized across multiple splits and 1 subsets.
 The dataset includes audio data.
 
 ## Dataset Structure
@@ -155,8 +115,6 @@ The dataset includes audio data.
 
 This dataset includes the following subsets:
 
-- **original**: 500 samples
-  - test: 500 samples
 - **denoised**: 500 samples
   - test: 500 samples
 
@@ -167,10 +125,10 @@ Load specific subset and split:
 from datasets import load_dataset
 
 # Load specific subset and split
-dataset = load_dataset('sujalappa/denoised-evaluation-dataset', 'original', split='train')
+dataset = load_dataset('sujalappa/denoised-evaluation-dataset', 'denoised', split='train')
 
 # Load all splits from a subset
-dataset = load_dataset('sujalappa/denoised-evaluation-dataset', 'original')
+dataset = load_dataset('sujalappa/denoised-evaluation-dataset', 'denoised')
 
 # Load everything
 dataset = load_dataset('sujalappa/denoised-evaluation-dataset')
@@ -178,7 +136,7 @@ dataset = load_dataset('sujalappa/denoised-evaluation-dataset')
 
 ## Dataset Creation
 
-This dataset was created using StreamableDatasetManager on 2025-09-28T13:50:54.974502.
+This dataset was created using StreamableDatasetManager on 2025-09-28T14:17:41.346363.
 
 ### Data Fields
 
@@ -198,11 +156,12 @@ The dataset includes the following columns:
 - **filename**: String data
 - **original_path**: String data
 - **audio**: Audio data (16kHz sampling rate)
+- **is_denoised**: Int64 data
 
 ## Technical Details
 
-- **Total samples**: 1,000
+- **Total samples**: 500
 - **Shard length**: 500
-- **Number of subsets**: 2
+- **Number of subsets**: 1
 - **Number of splits**: 2
 - **Audio format**: 16kHz sampling rate