Update README.md

README.md

@@ -46,7 +46,7 @@ It contains a large-scale training set (150k+ hours) and a curated evaluation se
 **Filtering rules:**
 - Only samples with successfully extracted word-level alignments are kept; failed alignments are skipped.
 - The average alignment confidence score is greater than a threshold **x**, where **x ∈ [0.2, 0.5]** depending on the source dataset.
-- Audio duration is between **2 and 30 seconds**, and the maximum pause between consecutive words does not exceed **4 seconds**.
+- Audio duration is between **0.5 and 30 seconds**, and the maximum pause between consecutive words does not exceed **4 seconds**.
 - Text is restricted to supported languages; samples containing characters outside the supported set (10 languages) are removed.
 - The text-to-duration ratio `len(txt) / dur` falls within a language-specific range.