Merge branch 'main' of hf.co:datasets/ttsds/listening_test_results

README.md

@@ -1,3 +1,31 @@
----
-license: mit
----
+---
+license: mit
+task_categories:
+- audio-classification
+language:
+- en
+tags:
+- MOS
+- CMOS
+- SMOS
+- synthetic
+- audio
+- speech
+- subjective
+- ratings
+- mos_prediction
+pretty_name: TTSDS Listening Test Results
+size_categories:
+- 10K<n<100K
+configs:
+  - config_name: default
+    data_files:
+      - split: train
+        path: subjective_results.csv
+---
+
+# Listening Test Results for TTSDS2
+
+This dataset contains all 11,000+ ratings collected for 20 synthetic speech systems for the TTSDS2 study (link coming soon).
+The scores are MOS (Mean Opinion Score), CMOS (Comparative Mean Opinion Score) and SMOS (Speaker Similarity Mean Opinion Score).
+All annotators included passed three attention checks throughout the survey.