Sync model card with upstream GitHub inference README

README.md

@@ -152,7 +152,7 @@ Stratified 150-sample eval (30 per scenario) on the
 We track DNSMOS P.808 (`sig_bak_ovr.onnx`) in TensorBoard but are deliberately
 *not* publishing OVRL numbers for this model. The scores we obtain (around 2.0
 overall, 2.1 on single-talk far-end) contradict informal listening —
-single-talk far-end with 52 dB of cancellation is audibly near-silent, not a
+single-talk far-end with ~48 dB of cancellation is audibly near-silent, not a
 "2-out-of-5" output. We suspect our DNSMOS invocation (input normalisation,
 silence handling, or ONNX model variant) is miscalibrated for AEC outputs
 and in particular for near-silent clips, which are out of distribution for a