From f6e22bb3b9f13d2bd17b23adc9655b49645665d7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Fri, 19 Jul 2024 11:42:13 -0700
Subject: [PATCH] processors(realtime-ai): add silero vad to the transport

---
 examples/foundational/18-realtime-ai.py         | 4 +++-
 src/pipecat/processors/frameworks/realtimeai.py | 7 +------
 2 files changed, 4 insertions(+), 7 deletions(-)

diff --git a/examples/foundational/18-realtime-ai.py b/examples/foundational/18-realtime-ai.py
index d00225a61..8a0a7683c 100644
--- a/examples/foundational/18-realtime-ai.py
+++ b/examples/foundational/18-realtime-ai.py
@@ -13,6 +13,7 @@ from pipecat.pipeline.task import PipelineParams, PipelineTask
 from pipecat.pipeline.runner import PipelineRunner
 from pipecat.processors.frameworks.realtimeai import RealtimeAIProcessor
 from pipecat.transports.services.daily import DailyParams, DailyTransport
+from pipecat.vad.silero import SileroVADAnalyzer
 
 from runner import configure
 
@@ -31,9 +32,10 @@ async def main(room_url, token):
         token,
         "Realtime AI",
         DailyParams(
-            audio_in_enabled=True,
             audio_out_enabled=True,
             transcription_enabled=True,
+            vad_enabled=True,
+            vad_analyzer=SileroVADAnalyzer()
         ))
 
     rtai = RealtimeAIProcessor(
diff --git a/src/pipecat/processors/frameworks/realtimeai.py b/src/pipecat/processors/frameworks/realtimeai.py
index 5aaf0e204..d296416fa 100644
--- a/src/pipecat/processors/frameworks/realtimeai.py
+++ b/src/pipecat/processors/frameworks/realtimeai.py
@@ -25,7 +25,6 @@ from pipecat.services.ai_services import AIService
 from pipecat.services.cartesia import CartesiaTTSService
 from pipecat.services.openai import OpenAILLMService
 from pipecat.transports.base_transport import BaseTransport
-from pipecat.vad.silero import SileroVAD
 
 DEFAULT_MESSAGES = [
     {
@@ -141,8 +140,6 @@ class RealtimeAIProcessor(FrameProcessor):
 
     async def _handle_setup(self, setup: RealtimeAISetup | None):
         try:
-            vad = SileroVAD()
-
             model = DEFAULT_MODEL
             if setup and setup.config and setup.config.llm and setup.config.llm.model:
                 model = setup.config.llm.model
@@ -166,7 +163,6 @@ class RealtimeAIProcessor(FrameProcessor):
             self._tts = self._tts_cls(api_key=self._tts_api_key, voice_id=voice)
 
             pipeline = Pipeline([
-                vad,
                 self._tma_in,
                 self._llm,
                 self._tts,
@@ -220,8 +216,7 @@ class RealtimeAIProcessor(FrameProcessor):
         # send any messages. So, we setup a super basic pipeline with just the
         # output transport so we can send messages.
         if not self._pipeline:
-            # We add the SilerVAD() so the audio doesn't go through.
-            pipeline = Pipeline([SileroVAD(), self._transport.output()])
+            pipeline = Pipeline([self._transport.output()])
             self._pipeline = pipeline
 
             parent = self.get_parent()