Add a face to your Azure Speech pipeline

LemonSlice puts a real-time interactive avatar on top of Azure AI Speech. Whether you use Azure TTS voices, transcription, or realtime voice, the avatar turns that audio into natural, face-to-face conversation with your users.

LemonSlice

At a glance

Categories
Voice (TTS), Transcription (STT + VAD), Realtime
Status
Compatible (works with any)
Type
Avatar API
Languages
Python, Node.js
Setup time
~5 minutes

How the integration works

Azure Speech keeps doing what it does today. LemonSlice turns the audio into a face.

  1. Keep your Azure Speech setup

    Your Azure AI Speech resources keep handling text to speech, speech to text, and realtime voice. Nothing changes in your subscription, regions, or voice configuration.

  2. Stream the audio to LemonSlice

    Pipe the synthesized audio from Azure Speech into a LemonSlice avatar session. The avatar renders lip sync and expression in real time as each chunk arrives.

  3. The avatar video streams to your users

    LemonSlice delivers the avatar video over WebRTC, natively via LiveKit, Daily (Pipecat), or Agora. Users see a responsive face in about the blink of an eye.

Add your real-time avatar in less than 5 minutes

A Azure Speech session with a LemonSlice avatar.

Choose a code example
from livekit import agents
from livekit.agents import AgentSession, RoomOutputOptions
from livekit.plugins import lemonslice

async def entrypoint(ctx: agents.JobContext):
    await ctx.connect()

    session = AgentSession(
        # Add Azure Speech for TTS and STT, plus your LLM, here
    )

    avatar = lemonslice.AvatarSession(
        # Publicly accessible image URL for the avatar
        agent_image_url="...",
        # Prompt to guide the avatar's demeanor while speaking
        agent_prompt="a person talking",
        # Recommended. Adapts video quality to each viewer's bandwidth
        simulcast=True,
    )

    # Start the avatar and wait for it to join
    session_id = await avatar.start(session, room=ctx.room)

    # Start your agent session with the user
    await session.start(
        # ... room, agent, room_options, etc....
    )

A real-time avatar that feels human

Control actions and emotions

The Action Engine brings face, hands, shoulders, and posture to life. 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 animates non-humans, a turtle or a smiling teacup, with the same ease as cartoons and photorealistic people.

Ultra-low latency

LemonSlice delivers the fastest responses of any major avatar provider, so every exchange feels instant.

Create an avatar from one image

From one photo comes a full 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 alongside Azure Speech run at production scale.

Thousands of concurrent calls

Simultaneous avatar sessions scale into the thousands, with high-volume production workloads as the design target.

Multi-region servers

Avatar sessions launch from the region closest to your users so latency stays at a minimum.

Zero data retention mode

Run entirely without stored conversation data, which suits privacy-sensitive enterprise workflows.

What builders ship with Azure Speech + LemonSlice

Create avatars for everything from customer support to corporate training, clinical simulation, and airport kiosks.

Customer support agent

Answer, triage, and resolve customer questions face to face.

Corporate training coach

Roleplay difficult conversations and train staff on policy.

Virtual patient simulator

Medical students practice diagnosis on a responsive patient.

Frequently asked questions

Keep your Azure AI Speech pipeline as is and stream its audio output into a LemonSlice avatar session through the API. The avatar video then streams to your users over WebRTC.
No. Azure Speech handles text to speech, speech to text, and realtime voice, while LemonSlice renders the visual layer, a real-time avatar with lip sync and expression.
Yes. LemonSlice animates whatever audio Azure Speech produces, so your chosen voices, languages, and speech settings carry over unchanged.
LemonSlice offers a zero data retention mode that runs avatar sessions without storing conversation data, which suits regulated enterprise workflows built on Azure.
All three lanes. Azure TTS voices, speech transcription, and realtime voice each stream into the avatar the same way.
LemonSlice offers a zero data retention mode and multi-region deployment, so avatar sessions can match the privacy posture of your Azure stack.

Give your Azure Speech pipeline a face

Keep your Azure stack, add a real-time avatar with a 5-minute API setup.