Text To Multimedia

Text to Multimedia Pipeline

Generate a complete multimedia package — image, video, and audio — from a single text description using each::sense — the intelligent AI agent that automatically selects the best model for your request.

Quick Start

Requires an each::labs API key. Get one at eachlabs.ai.

Step 1: Generate the Image

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "messages": [
      {
        "role": "user",
        "content": "A cozy cabin in a snowy forest at night, warm golden light glowing from the windows, smoke rising from the stone chimney, northern lights dancing in the sky above, tall snow-covered pines surrounding the cabin, photorealistic winter landscape"
      }
    ],
    "stream": false
  }'

Step 2: Generate the Video

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "messages": [
      {
        "role": "user",
        "content": [
              {"type": "text", "text": "Animate this winter cabin scene: smoke gently curling from the chimney, northern lights slowly shifting colors and dancing across the sky, snowflakes drifting down softly, a warm flicker of firelight in the windows, slow dolly-in camera movement, peaceful and atmospheric"}
            ]
          }
    ],
    "stream": false
  }'

Step 3: Generate the Audio

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "messages": [
      {
        "role": "user",
        "content": "Generate ambient audio for a winter cabin scene: gentle wind blowing through pine trees, distant crackling of a fireplace, soft snowfall sounds, occasional owl hooting in the distance, peaceful winter night atmosphere"
      }
    ],
    "stream": false
  }'

Full Pipeline in Python

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_EACHLABS_API_KEY",
    base_url="https://eachsense-agent.core.eachlabs.run/v1"
)

description = "A cozy cabin in a snowy forest at night with northern lights"

# Step 1: Generate image
print("Step 1: Generating image...")
image_response = client.chat.completions.create(
    model="eachsense/beta",
    messages=[{
        "role": "user",
        "content": f"{description}, warm golden light from windows, smoke from chimney, snow-covered pines, photorealistic"
    }]
)
print("Image:", image_response.choices[0].message.content)

# Step 2: Generate video (using image URL from Step 1)
print("Step 2: Generating video...")
video_response = client.chat.completions.create(
    model="eachsense/beta",
    messages=[{
        "role": "user",
        "content": "Animate this scene: northern lights dancing, smoke curling from chimney, snowflakes drifting, slow cinematic camera movement"
    }],
    # Images are included in the message content array above
)
print("Video:", video_response.choices[0].message.content)

# Step 3: Generate audio
print("Step 3: Generating audio...")
audio_response = client.chat.completions.create(
    model="eachsense/beta",
    messages=[{
        "role": "user",
        "content": "Generate ambient winter night audio: gentle wind in pine trees, crackling fireplace, soft snowfall, occasional owl sounds"
    }]
)
print("Audio:", audio_response.choices[0].message.content)

Streaming

Set "stream": true for real-time SSE responses, or "stream": false for complete result in a single response. Streaming is useful for showing progress in UIs; non-streaming is simpler for scripts and automation.

Tips for Multimedia Generation

• Start with a strong visual description — the image sets the tone for all subsequent media.
• Adapt prompts for each format — images need composition details, videos need motion, audio needs soundscape.
• Keep the theme consistent — use the same setting, mood, and atmosphere across all three prompts.
• Generate audio that matches the visual mood — calm visuals with calm audio, dramatic visuals with intense audio.
• Process sequentially — image first (for video reference), then video, then audio (can be parallel with video).

Examples

Cinematic Nature Scene

Image:

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "messages": [{"role": "user", "content": "A dramatic thunderstorm over a vast wheat field, dark purple storm clouds with lightning bolts, golden wheat swaying in strong wind, a lone farmhouse in the distance, cinematic wide angle, photorealistic"}],
    "stream": false
  }'

Video:

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "messages": [{"role": "user", "content": [
              {"type": "text", "text": "Animate this storm scene: wheat field swaying dramatically in gusting wind, lightning flash illuminating the clouds, rain beginning to fall, camera slowly tracking forward toward the farmhouse, cinematic stormy atmosphere"}
            ]
          }
    ],
    "stream": false
  }'

Audio:

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "messages": [{"role": "user", "content": "Generate storm soundscape audio: rolling thunder in the distance, strong wind gusts through grain fields, scattered raindrops building to a downpour, occasional close thunder cracks, dramatic atmospheric storm sounds"}],
    "stream": false
  }'

Workflow: Text to Multimedia

1. Write the creative brief — One description covering the scene, mood, and purpose.
2. Generate the image — Create the visual anchor for the multimedia package.
3. Animate to video — Use the image as reference for video generation.
4. Generate matching audio — Create ambient sound or music that fits the scene.
5. Combine in editor — Layer video and audio together, add titles if needed.
6. Export for platforms — Render in the appropriate formats for target distribution.

Related Skills

• Image to Video Pipeline — Image-to-video animation
• Product Media Suite — Product-focused multimedia
• Brand Asset Pipeline — Brand identity multi-asset generation
• Text to Speech — Narration audio generation

Related Models

• flux-2-max — High-quality image generation
• seedream-v4-5 — Creative multimedia styles
• nano-banana-pro — Fast pipeline execution

Documentation

• each::sense Overview
• each::labs API