Update README.md

README.md

@@ -124,9 +124,4 @@ Context-dependent turn-taking (400 held-out samples)
 Single-turn datasets (Orpheus aggregate)
 | **Dataset** | **UltraVAD** | **Smart-Turn V2** |
 |---|---:|---:|
-| **orpheus-aggregate-train** | 93.7% | N/A |
-| **orpheus-aggregate-test** | N/A | 94.3% |
-
-Notes:
-Smart-Turn V2 test scores are reported from their [repo](https://huggingface.co/pipecat-ai/smart-turn-v2); UltraVAD uses their train splits for comparison due to test set unavailability. The aggregate numbers are within ~1 percentile, suggesting comparability.
-We use both recommended thresholds (0.1 for ultraVAD and 0.5 for smart-turnv2).
+| **orpheus-aggregate-test** | 93.7% | 94.3% |