Add a face to your Deepgram pipeline

LemonSlice puts a lifelike, real-time avatar on top of Deepgram's voice stack. Whether you run Nova STT, Aura TTS, or a full voice pipeline, the interactive avatar turns that audio into natural, face-to-face conversation with your customers.

LemonSlice

At a glance

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

How the integration works

Deepgram (TTS) keeps doing what it does today. LemonSlice turns the audio into a face.

  1. Keep your Deepgram setup

    Your Nova transcription and Aura voice synthesis keep running unchanged. No model swaps, no pipeline rewrites, no new configuration on the Deepgram side.

  2. Stream the audio to LemonSlice

    Pipe the Aura TTS output into a LemonSlice avatar session through the API. LemonSlice renders the avatar in real time with lip sync and expression matched to the audio.

  3. The avatar video streams to your users

    LemonSlice delivers the avatar video over WebRTC, natively via LiveKit, Daily (Pipecat), or Agora, so it drops into the same real-time infrastructure your voice pipeline already uses.

Drop in your real-time avatar with a short snippet

A Deepgram (TTS) 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 Deepgram 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....
    )

An interactive avatar that feels human

Control actions and emotions

The Action Engine animates the whole performance, face, hands, shoulders, and posture. You control actions and emotions through the API, down to a prompt like "talk with hand gestures."

Works with any character, in any style

LemonSlice stands alone in animating non-humans (a smiling teacup, a turtle) as easily as photorealistic people and cartoons.

Ultra-low latency

LemonSlice delivers the quickest response time among major avatar providers, keeping conversations feeling instant.

Create an avatar from one image

An interactive avatar starts from one photo. No training or fine-tuning required. Create unlimited custom characters with real-time outfit or setting changes.

Built for enterprise scale

Avatar sessions alongside Deepgram (TTS) run at production scale.

Thousands of concurrent calls

Run thousands of avatar sessions side by side, designed for high-volume production workloads.

Multi-region servers

Launch avatar sessions from the region closest to your users to keep latency low.

Zero data retention mode

Run avatar sessions that store no conversation data, designed for privacy-sensitive enterprise workflows.

What builders ship with Deepgram + LemonSlice

Create avatars for everything from language practice to customer support, live streaming, and corporate training.

Language practice partner

Practice real conversations in another language.

Customer support agent

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

Streaming host

A live host that reads and responds to chat.

Frequently asked questions

Keep your Nova STT and Aura TTS setup, then stream the Aura audio output into a LemonSlice avatar session through the API. The avatar video streams back to your users over WebRTC.
No. Deepgram keeps handling transcription and voice synthesis while LemonSlice adds the visual layer, a real-time avatar rendered live from that audio.
Yes. LemonSlice animates whatever audio you stream in, so the avatar speaks with exactly the Aura voice you chose.
Yes. Nova is the most common STT pairing in avatar stacks, and your transcription flow stays untouched because LemonSlice only consumes the audio side of the pipeline.
Yes. LemonSlice lip syncs to the audio itself rather than to any specific language, so the interactive avatar handles whatever language your Deepgram voice pipeline produces.
LemonSlice has the fastest response time of any major avatar provider, so conversations keep their natural rhythm. Multi-region servers also deploy sessions near your users to keep every exchange feeling immediate.

Give your Deepgram pipeline a face

Stream Aura audio to a live avatar with a 5-minute API setup.