Replace the train split with the 40 timbre-vector clips

Vietnamese text from gemini-3.8-flash, speech from VieNeu-TTS conditioned on one 192-d timbre vector from each English clip.

README.md

@@ -16,10 +16,10 @@ dataset_info:
         decode: false
   splits:
   - name: train
-    num_bytes: 230667
-    num_examples: 10
-  download_size: 233983
-  dataset_size: 230667
+    num_bytes: 896290
+    num_examples: 40
+  download_size: 899011
+  dataset_size: 896290
 configs:
 - config_name: default
   data_files:

data/train-00000-of-00001.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a3b77affff320efb67f2e5e7af63a05a4281db2faabfcc1da241f31b78f2e1a
-size 233983
+oid sha256:f4ff2cb99535ff08bc191249cbdd1602a619fc47dd26c57f5a36776d71e59b68
+size 899011