Please use this identifier to cite or link to this item:
http://arks.princeton.edu/ark:/88435/dsp018p58ph28g
Title: | Applications and ethics of a voice-activated robot that writes in a user's own handwriting |
Authors: | Narayanan, Rohit |
Advisors: | Houck, Andrew |
Department: | Electrical and Computer Engineering |
Certificate Program: | Program in Values and Public Life Robotics & Intelligent Systems Program |
Class Year: | 2024 |
Abstract: | For people who have lost the ability to write by hand, an automated hands-free device that writes a letter or note in their handwriting can help them reclaim the skill. In this study, we design a system to scan text from handwriting samples and generate a handwriting style. The system takes a voice input of a prompt for a letter to be written and feeds it to a Large Language Model (LLM) to compose a letter iteratively with the user until the user is satisfied. The system then writes that letter in the user’s handwriting on a sheet of paper via the AxiDraw V3 robot. The system was successfully able to mimic human handwriting – in some cases evaluators confused handwriting written by the machine and human handwriting. The system also accurately reproduced different handwriting styles and improved the accuracy of characters in a letter using a novel Character Refinement algorithm. This study is unique in considering some of the thorny ethical questions that come with such a device. Given the unique risks of using the robot for forgery, we implement authentication through voice print and a spoken password. |
URI: | http://arks.princeton.edu/ark:/88435/dsp018p58ph28g |
Type of Material: | Princeton University Senior Theses |
Language: | en |
Appears in Collections: | Electrical and Computer Engineering, 1932-2024 Robotics and Intelligent Systems Program |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
NARAYANAN-ROHIT-THESIS.pdf | 1.94 MB | Adobe PDF | Request a copy |
Items in Dataspace are protected by copyright, with all rights reserved, unless otherwise indicated.