Update README.md

README.md

@@ -6,26 +6,30 @@ size_categories:
 - 10M<n<100M
 ---
 
-Keys2Paragraphs.txt (internal name in scripts: yago_wiki.txt): 
-    4.3 million unique words/phrases (English Wikipedia titles or their parts) occurring in 33.8 million English Wikipedia paragraphs.
+This is the dataset presented in my [ASRU-2023 paper](https://arxiv.org/abs/2309.17267).
 
-Keys2Corruptions.txt (internal name in scripts: sub_misspells.txt):
-    26 million phrase pairs in the corrupted phrase inventory, as recognized by different ASR models
+It consists of multiple files:
 
-Keys2Related.txt (internal name in scripts: related_phrases.txt):
-    62.7 million phrase pairs in the related phrase inventory
+    Keys2Paragraphs.txt (internal name in scripts: yago_wiki.txt): 
+        4.3 million unique words/phrases (English Wikipedia titles or their parts) occurring in 33.8 million English Wikipedia paragraphs.
 
-FalsePositives.txt (internal name in scripts: false_positives.txt):
-    449 thousand phrase pairs in the false positive phrase inventory
+    Keys2Corruptions.txt (internal name in scripts: sub_misspells.txt):
+        26 million phrase pairs in the corrupted phrase inventory, as recognized by different ASR models
 
-NgramMappings (internal name in scripts: replacement_vocab_filt.txt):
-    5.5 million character n-gram mappings dictionary
+    Keys2Related.txt (internal name in scripts: related_phrases.txt):
+        62.7 million phrase pairs in the related phrase inventory
 
-asr
-    outputs of g2p+tts+asr using 4 different ASR systems (conformer ctc was used twice),
-    gives pairs of initial phrase and its recognition result.
-    Does not include .wav files, but these can be reproduced by feeding g2p to tts 
+    FalsePositives.txt (internal name in scripts: false_positives.txt):
+        449 thousand phrase pairs in the false positive phrase inventory
 
-giza
-    raw outputs of GIZA++ alignments for each corpus,
-    from these we get NgramMappings.txt and Keys2Corruptions.txt
+    NgramMappings (internal name in scripts: replacement_vocab_filt.txt):
+        5.5 million character n-gram mappings dictionary
+
+    asr
+        outputs of g2p+tts+asr using 4 different ASR systems (conformer ctc was used twice),
+        gives pairs of initial phrase and its recognition result.
+        Does not include .wav files, but these can be reproduced by feeding g2p to tts 
+
+    giza
+        raw outputs of GIZA++ alignments for each corpus,
+        from these we get NgramMappings.txt and Keys2Corruptions.txt