Update batch_confirm_hf.py to load the entire validated dataset instead of a subset.

2025-08-02 12:24:27 +03:30
parent d16f4a84bb
commit 3cc3bf0d3e
1 changed files with 1 additions and 1 deletions
--- a/vosk/test_files/batch_confirm_hf.py
+++ b/vosk/test_files/batch_confirm_hf.py
@@ -14,7 +14,7 @@ from huggingface_hub import HfApi, create_repo
 print("Loading dataset...")
 ds = load_dataset(
    "Ashegh-Sad-Warrior/Persian_Common_Voice_17_0",
-    split="validated[:5]",
+    split="validated",
    streaming=False
 ).cast_column("audio", Audio(sampling_rate=16000))