Bytedance Avatar Omni Human 1.5 API Documentation

Bytedance Avatar Omni Human 1.5 API Documentation

Playground

OmniHuman 1.5 converts audio and visual cues into lifelike avatar animations for virtual humans, storytelling, and interactive agents. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Avatar Omni-Human 1.5 is an advanced vision-audio fusion model designed to animate avatars through cognitive and emotional simulation. By combining image and audio inputs, it brings static portraits to life — generating natural facial expressions, synchronized lip movements, and realistic emotional responses.

🧠 Concept

Inspired by the paper “Instilling an Active Mind in Avatars via Cognitive Simulation”, the model simulates attention, emotion, and cognition to create avatars that don’t just move — they react intelligently.

🌟 Key Features

• Audio-Driven Realism Generates precise lip-sync and emotional nuance directly from voice input.

• Expressive Cognitive Simulation Models subtle eye movements, micro-expressions, and reactive behavior to emulate human presence.

• Universal Avatar Adaptation Works with any static portrait or illustration to create consistent, lifelike performance.

• Cross-Domain Support Handles both photorealistic and stylized avatars, adapting its realism to the visual style.

• Flexible Output Encoding Choose between URL output or BASE64 encoding for seamless integration via API.

⚙️ Parameters

💰 Pricing

💡 Use Cases

• Digital Avatars & VTubing — Drive realistic avatars from real voices in real time.
• Virtual Humans & NPCs — Give game or metaverse characters believable cognitive reactions.
• Marketing & Storytelling — Create expressive digital spokespeople or narrators.
• AI Companions & Education — Build avatars that engage naturally in learning or dialogue contexts.

📝 Notes

• The longer the audio, the higher the total cost (calculated per second).
• For best results, use clear, high-quality audio and well-lit frontal images.
• BASE64 output is API-only, useful for direct embedding into web applications.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "image": "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg",
  "audio": "https://interactive-examples.mdn.mozilla.net/media/cc0-audio/t-rex-roar.mp3"
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/bytedance/avatar-omni-human-1.5" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL=$(printf '%s' "${TASK}" | jq -r '.urls.get // empty')
if [ -z "${RESULT_URL}" ]; then RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"; fi

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    created|processing) sleep 2 ;;
    *) printf 'Unexpected status: %s
' "${STATUS}" >&2; exit 1 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters