PodCastle and songle: Crowdsourcing-based web services for spoken document retrieval and active music listening
Masataka Goto, Jun Ogata, Kazuyoshi Yoshii, Hiromasa Fujihara, Matthias Mauch, Tomoyasu Nakano · 2012
In this talk, we describe two web services for speech and music, PodCastle (Figure 1) and Songle (Figure 2), that collect and amplify voluntary contributions by anonymous users to improve user experiences. Our goal is to provide users with public web services based on speech recognition, music understanding, signal processing, machine learning, and crowdsourcing so that they can experience state-of-the-art research-level technologies through useful services.