Switch noise cancellation to ai-coustics QUAIL_VF_L

Use ai-coustics Voice Focus for non-SIP participants, which has
significantly lower word error rate (11.8%) compared to Krisp BVC
(23.5%). Keeps Krisp BVCTelephony for SIP participants.

Author: Topherhindman

pyproject.toml

@@ -10,6 +10,7 @@ requires-python = ">=3.10, <3.15"
 
 dependencies = [
     "livekit-agents[silero,turn-detector]~=1.4",
+    "livekit-plugins-ai-coustics",
     "livekit-plugins-noise-cancellation~=0.2",
     "python-dotenv",
 ]

src/agent.py

@@ -12,7 +12,7 @@
     inference,
     room_io,
 )
-from livekit.plugins import noise_cancellation, silero
+from livekit.plugins import ai_coustics, noise_cancellation, silero
 from livekit.plugins.turn_detector.multilingual import MultilingualModel
 
 logger = logging.getLogger("agent")
@@ -115,7 +115,9 @@ async def my_agent(ctx: JobContext):
                     noise_cancellation.BVCTelephony()
                     if params.participant.kind
                     == rtc.ParticipantKind.PARTICIPANT_KIND_SIP
-                    else noise_cancellation.BVC()
+                    else ai_coustics.audio_enhancement(
+                        model=ai_coustics.EnhancerModel.QUAIL_VF_L
+                    )
                 ),
             ),
         ),