diff --git a/changelog/3927.added.md b/changelog/3927.added.md
new file mode 100644
index 000000000..378815fc3
--- /dev/null
+++ b/changelog/3927.added.md
@@ -0,0 +1 @@
+- Added `vad_threshold` parameter to `AssemblyAIConnectionParams` for configuring voice activity detection sensitivity in U3 Pro. Aligning this with external VAD thresholds (e.g., Silero VAD) prevents the "dead zone" where AssemblyAI transcribes speech that VAD hasn't detected yet.
diff --git a/changelog/3927.changed.md b/changelog/3927.changed.md
new file mode 100644
index 000000000..b397e86a2
--- /dev/null
+++ b/changelog/3927.changed.md
@@ -0,0 +1 @@
+- ⚠️ Removed `formatted_finals` and `word_finalization_max_wait_time` from `AssemblyAIConnectionParams` as these were v2 API parameters not supported in v3. Clarified that `format_turns` only applies to Universal-Streaming models; U3 Pro has automatic formatting built-in.