Speech to Reality: On-Demand Production using Natural Language, 3D Generative AI, and Discrete Robotic Assembly

2024-09-27

arXiv.org

AI Summary

This study addresses core challenges in physical manufacturing with generative AI—namely variability, structural integrity, efficiency, and material waste—enabling zero-expertise users to perform end-to-end autonomous fabrication via voice commands. We propose a novel “voice-driven + voxelized discrete assembly” paradigm: integrating automatic speech recognition (ASR), large language models (LLMs), and 3D diffusion models to generate manufacturable 3D structures; discretizing them into lattice-based voxel units; and optimizing geometry-aware assembly sequences with motion planning for robust execution on a six-axis robotic arm. The approach eliminates conventional CAD modeling and programming requirements. In under five minutes, it completes the full pipeline—from voice input to AI generation to robotic construction—for diverse objects (e.g., chairs, shelving units). Experimental validation confirms structural load-bearing compliance, sub-0.5 mm assembly accuracy, and substantial improvements in generation controllability, real-time manufacturability, and material utilization efficiency.

Technology Category

• Intelligent Robots: Embodied AI
• Natural Language Processing: Generation
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

We present a system that transforms speech into physical objects by combining 3D generative Artificial Intelligence with robotic assembly. The system leverages natural language input to make design and manufacturing more accessible, enabling individuals without expertise in 3D modeling or robotic programming to create physical objects. We propose utilizing discrete robotic assembly of lattice-based voxel components to address the challenges of using generative AI outputs in physical production, such as design variability, fabrication speed, structural integrity, and material waste. The system interprets speech to generate 3D objects, discretizes them into voxel components, computes an optimized assembly sequence, and generates a robotic toolpath. The results are demonstrated through the assembly of various objects, ranging from chairs to shelves, which are prompted via speech and realized within 5 minutes using a 6-axis robotic arm.

Problem

• Transforms speech into physical objects using AI and robotics
• Enables non-experts to create objects without 3D modeling skills
• Addresses challenges like design variability and structural integrity

Innovation

• Combines 3D generative AI with robotic assembly
• Uses natural language for accessible design
• Employs discrete voxel assembly for structural integrity