dataset card established

README.md

@@ -1,3 +1,32 @@
+---
+# dataset card
+# https://github.com/huggingface/hub-docs/blob/main/datasetcard.md?plain=1
+
+language: 
+- en
+language_details:
+- en-US
+
+pretty_name: "UCLA Speaker Variability Database - subset for SDialog"
+tags:
+- audio
+- speech
+arxiv: 2008.03616
+license: "any valid license identifier"
+task_categories:
+# see list https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/src/pipelines.ts
+- text-to-speech
+- speaker-identification
+
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: "audio/*.wav"
+
+
+---
+
 # Speakers voice samples from The UCLA Speaker Variability Database