The City of Toronto recently combined six former communities to create one megacity. With an annual budget of $8.5 billion (CDN), 2.1 million population, and over 50,000 employees, the task of identifying assets, their condition, and location was overwhelming. The City of Toronto selected a speech recognition application combined with GPS to locate the estimated 1.2 million street signs, 160,000 streetlights, 5,100 kilometers of roads, and other City assets. Speech recognition has been successfully used in a pilot project with the City. The system is being implemented and is solving the City of Toronto's data problem. The paper will highlight the successes and lessons learned from using this cutting-edge technology to build and maintain Toronto's ArcView GIS and ArcInfo maps.