how do i add noise cancellation for telephony to this model LiveKit Community #ask-ai

how do i add noise cancellation for telephony to t...

strong-pizza-31661

07/21/2025, 7:02 PM

how do i add noise cancellation for telephony to this? model = openai.realtime.RealtimeModel( voice="shimmer", temperature=0.8, #model="gpt-4o-realtime-preview-2025-06-03", # Tell Whisper to expect Turkish (“tr”) audio _input_audio_transcription_=None, _turn_detection_=None, modalities=["text"] ) # Create AgentSession with the multimodal model session = AgentSession( llm=model, tts=elevenlabs.TTS( language=LANG_MODE, _voice_id_="cgSgspJ2msm6clMCkdW9", _streaming_latency_=3, ), vad=silero.VAD.load(), stt = deepgram.STT( model="nova-2-general", language=LANG_MODE, _sample_rate_=8000, # critical ), _#turn_detection=MultilingualModel(),_ )

tall-airline-16756

07/21/2025, 7:02 PM

Thanks for reaching out, Rhythm. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.

tall-airline-16756

07/21/2025, 7:03 PM

To add noise cancellation optimized for telephony audio (such as 8kHz streams) in your pipeline, use LiveKit's

BVCTelephony

noise cancellation model. This is specifically designed for telephony use cases and works well with STT, TTS, and VAD components. Installation:

Copy code

pip install "livekit-plugins-noise-cancellation~=0.2"

Integration (Agent-Level, recommended for your setup):

Copy code

from livekit.plugins import noise_cancellation

await session.start(
    # ...,
    room_input_options=room_io.RoomInputOptions(
        noise_cancellation=noise_cancellation.BVCTelephony(),
    ),
)

Alternative: AudioStream-Level Integration

Copy code

stream = rtc.AudioStream.from_track(
    track=track,
    noise_cancellation=noise_cancellation.BVCTelephony(),
)

• Use

noise_cancellation.BVCTelephony()

for best results with telephony audio. • Disable any client-side noise cancellation to avoid double processing. • This feature is available for LiveKit Cloud customers. The

BVCTelephony

model is optimized for telephony and will provide optimal noise suppression for your use case.

Open in Slack

Previous Next