Add cached elevenlabs in service factory

api/services/pipecat/service_factory.py

@@ -2,7 +2,7 @@ from typing import TYPE_CHECKING
 
 from fastapi import HTTPException
 
-from api.constants import MPS_API_URL
+from api.constants import MPS_API_URL, REDIS_URL
 from api.services.configuration.registry import ServiceProviders
 from pipecat.services.azure.llm import AzureLLMService
 from pipecat.services.cartesia.stt import CartesiaSTTService
@@ -11,6 +11,7 @@ from pipecat.services.deepgram.tts import DeepgramTTSService
 from pipecat.services.dograh.llm import DograhLLMService
 from pipecat.services.dograh.stt import DograhSTTService
 from pipecat.services.dograh.tts import DograhTTSService
+from pipecat.services.elevenlabs.elevenlabs_cached_tts import ElevenLabsCachedTTSService
 from pipecat.services.elevenlabs.tts import ElevenLabsTTSService
 from pipecat.services.google.llm import GoogleLLMService
 from pipecat.services.groq.llm import GroqLLMService
@@ -82,7 +83,7 @@ def create_tts_service(user_config, audio_config: "AudioConfig"):
         )
     elif user_config.tts.provider == ServiceProviders.ELEVENLABS.value:
         voice_id = user_config.tts.voice.split(" - ")[1]
-        return ElevenLabsTTSService(
+        return ElevenLabsCachedTTSService(
             reconnect_on_error=False,
             api_key=user_config.tts.api_key,
             voice_id=voice_id,
@@ -91,6 +92,7 @@ def create_tts_service(user_config, audio_config: "AudioConfig"):
                 stability=0.8, speed=user_config.tts.speed, similarity_boost=0.75
             ),
             text_filters=[xml_function_tag_filter],
+            cache_redis_url=REDIS_URL,
         )
     elif user_config.tts.provider == ServiceProviders.DOGRAH.value:
         # Convert HTTP URL to WebSocket URL for TTS
@@ -102,6 +104,8 @@ def create_tts_service(user_config, audio_config: "AudioConfig"):
             model=user_config.tts.model.value,
             voice=user_config.tts.voice.value,
             text_filters=[xml_function_tag_filter],
+            cache_enabled=True,
+            redis_url=REDIS_URL,
         )
     else:
         raise HTTPException(

api/services/pipecat/transport_setup.py

@@ -19,7 +19,6 @@ from pipecat.audio.mixers.soundfile_mixer import SoundfileMixer
 from pipecat.audio.turn.smart_turn.base_smart_turn import SmartTurnParams
 from pipecat.audio.turn.smart_turn.local_smart_turn_v3 import LocalSmartTurnAnalyzerV3
 from pipecat.audio.vad.silero import SileroVADAnalyzer, VADParams
-from pipecat.serializers.plivo import PlivoFrameSerializer
 from pipecat.serializers.twilio import TwilioFrameSerializer
 from pipecat.serializers.vobiz import VobizFrameSerializer
 from pipecat.serializers.vonage import VonageFrameSerializer

api/services/telephony/providers/vobiz_provider.py

@@ -299,11 +299,11 @@ class VobizProvider(TelephonyProvider):
         message handling to VobizFrameSerializer.
         """
         from api.services.pipecat.run_pipeline import run_pipeline_vobiz
-        
+
         first_msg = await websocket.receive_text()
         start_msg = json.loads(first_msg)
         logger.debug(f"Received the first message: {start_msg}")
-        
+
         # Validate that this is a start event
         if start_msg.get("event") != "start":
             logger.error(f"Expected 'start' event, got: {start_msg.get('event')}")
@@ -317,7 +317,7 @@ class VobizProvider(TelephonyProvider):
             start_data = start_msg.get("start", {})
             stream_id = start_data.get("streamId")
             call_id = start_data.get("callId")
-            
+
             if not stream_id or not call_id:
                 logger.error(f"Missing streamId or callId in start event: {start_data}")
                 await websocket.close(code=4400, reason="Missing streamId or callId")