Design and Implementation of Voice Interface Based on Google Cloud Platform

Dong-Wook Choi · Ji-Hoon Lee · Nammee Moon — 2nd author·KIBME Fall Conference 2019 (Capstone Design)2019Conference

PDF ↓

Designed and implemented a voice interface on Google Cloud Platform’s STT (Speech To Text) and TTS (Text To Speech) APIs.

  • Commands: read · write · run · quit · search
  • Search is handled by JSoup and Selenium crawling
  • Microphone input → STT conversion → a Java system interprets the command and controls the PC
  • The result is converted back to MP3 through TTS and played through the speaker

Built on Eclipse · GCP · MySQL, with a database for login and program path configuration.