diff --git a/livekit-agents/livekit/agents/multimodal/agent_playout.py b/livekit-agents/livekit/agents/multimodal/agent_playout.py
index a01d5b0ea..4ce66cae1 100644
--- a/livekit-agents/livekit/agents/multimodal/agent_playout.py
+++ b/livekit-agents/livekit/agents/multimodal/agent_playout.py
@@ -74,8 +74,8 @@ def __init__(
         self,
         *,
         audio_source: rtc.AudioSource,
-        stt: stt.STT,
-        stt_forwarder: stt.STTForwarder,
+        stt: stt.STT | None,
+        stt_forwarder: transcription.STTSegmentsForwarder | None,
     ) -> None:
         super().__init__()
         self._source = audio_source
@@ -180,7 +180,7 @@ async def _stt_stream_co() -> None:
             done, _ = await asyncio.wait(
                 [asyncio.gather(*tasks), handle._int_fut],
                 return_when=asyncio.FIRST_COMPLETED,
-            )
+            )  # type: ignore
 
             handle._total_played_time = (
                 handle._pushed_duration - self._source.queued_duration
diff --git a/livekit-agents/livekit/agents/multimodal/multimodal_agent.py b/livekit-agents/livekit/agents/multimodal/multimodal_agent.py
index afc65cdee..2d24cfd85 100644
--- a/livekit-agents/livekit/agents/multimodal/multimodal_agent.py
+++ b/livekit-agents/livekit/agents/multimodal/multimodal_agent.py
@@ -179,7 +179,7 @@ def __init__(
             transcription=transcription,
         )
 
-        if stt is not None and not stt.capabilities.streaming:
+        if stt and vad and not stt.capabilities.streaming:
             from .. import stt as speech_to_text
 
             stt = speech_to_text.StreamAdapter(
@@ -488,7 +488,7 @@ def _subscribe_to_microphone(self, *args, **kwargs) -> None:
                 self._subscribed_track = publication.track  # type: ignore
                 stream_24khz = rtc.AudioStream(
                     self._subscribed_track, sample_rate=24000, num_channels=1
-                )
+                )  # type: ignore
                 self._stt_forwarder = STTSegmentsForwarder(
                     room=self._room,
                     participant=self._linked_participant,
diff --git a/livekit-plugins/livekit-plugins-google/livekit/plugins/google/beta/realtime/stt.py b/livekit-plugins/livekit-plugins-google/livekit/plugins/google/beta/realtime/stt.py
index 9153039a7..4b791f408 100644
--- a/livekit-plugins/livekit-plugins-google/livekit/plugins/google/beta/realtime/stt.py
+++ b/livekit-plugins/livekit-plugins-google/livekit/plugins/google/beta/realtime/stt.py
@@ -26,8 +26,8 @@
     utils,
 )
 
-from google import genai
-from google.genai import types
+from google import genai  # type: ignore
+from google.genai import types  # type: ignore
 
 from .api_proto import LiveAPIModels