Add a face to your Pipecat pipeline

LemonSlice puts a lifelike, real-time avatar on top of Daily's Pipecat framework. Whether you run voice agents, multimodal pipelines, or Daily rooms, the real-time avatar turns that audio into natural, face-to-face conversation with your customers.

LemonSlice

At a glance

Category
Orchestration
Status
Live integration partner
Type
Avatar API
Languages
Python, Node.js
Setup time
~5 minutes

How the integration works

Daily / Pipecat keeps doing what it does today. LemonSlice turns the audio into a face.

  1. Keep your Pipecat pipeline as it is

    Your Pipecat agent keeps handling transcription, the LLM turn, and speech synthesis with the services you already use. No processors need to be rewritten and no logic moves.

  2. Stream the audio to LemonSlice

    Add the LemonSlice service to your pipeline. Your agent's synthesized speech streams to the LemonSlice avatar API, which renders lip sync and expression in real time.

  3. The avatar video streams to your users

    LemonSlice publishes the avatar video into your Daily room over WebRTC. Participants see and hear the avatar through the same Daily transport your Pipecat agent already uses, with LiveKit and Agora available as alternate delivery paths.

Add your real-time avatar with a few lines of code

A Daily / Pipecat session with a LemonSlice avatar.

Choose a code example
import os
import aiohttp
from pipecat.transports.lemonslice.transport import (
    LemonSliceNewSessionRequest,
    LemonSliceParams,
    LemonSliceTransport,
)

async def main():
    async with aiohttp.ClientSession() as session:
        transport = LemonSliceTransport(
            bot_name="Pipecat",
            api_key=os.getenv("LEMONSLICE_API_KEY"),
            session=session,
            session_request=LemonSliceNewSessionRequest(
                agent_image_url="...",
            ),
        )

        # stt, tts, llm...

An avatar that listens, reacts, and feels human

Control actions and emotions

Face, hands, shoulders, and posture all move through the Action Engine. You control actions and emotions directly through the API, down to a prompt like "talk with hand gestures."

Works with any character, in any style

LemonSlice is the only interactive avatar model that handles non-humans (a turtle, a smiling teacup) as easily as cartoons and photorealistic people.

Ultra-low latency

LemonSlice delivers the fastest response time of any major avatar provider, so conversations feel instant.

Create an avatar from one image

Any photo becomes an interactive avatar. No training or fine-tuning required. Create unlimited interactive avatars with real-time outfit or setting changes.

Built for enterprise scale

Avatar sessions in your Daily / Pipecat rooms run at production scale.

Thousands of concurrent calls

Run avatar sessions by the thousands at once, built for high-volume production workloads.

Multi-region servers

Serve avatar sessions from the region nearest your users for lowest latency.

Zero data retention mode

Run sessions with zero stored conversation data, built for privacy-sensitive enterprise workflows.

What builders ship with Daily / Pipecat + LemonSlice

Create avatars for everything from airport kiosks to interviewing, product demos, and corporate training.

Airport help desk

Wayfinding and flight answers on a kiosk screen.

Interviewer

Structured screening interviews, face to face.

Product demo agent

Walk prospects through the product and answer questions live.

Frequently asked questions

Add the LemonSlice service to your pipeline and point it at an avatar image. Your agent's audio streams to the LemonSlice API and the rendered avatar video publishes into your Daily room.
No. Daily handles the WebRTC transport, Pipecat orchestrates your agent pipeline, and LemonSlice renders the real-time avatar layer on top.
Yes. LemonSlice sits downstream of your pipeline and animates whatever audio your agent produces, so every service you have configured stays in place.
No. The avatar joins the Daily room as a video participant over WebRTC, so it appears in any client that already connects to your rooms.
Yes. LemonSlice publishes avatar video over WebRTC, with LiveKit and Agora available as alternate delivery paths, so your Pipecat agent is not locked to one transport.
One photo is all it takes. LemonSlice builds an interactive avatar from a single image with no training or rigging, and it handles photorealistic people, cartoons, and non-human characters alike.

Give your Pipecat agent a face

5-minute API setup. Your Daily pipeline stays exactly as it is.