An Auditory Model Based Transcriber of Vocal Queries
Tom De Mulder, Jean‐Pierre Martens, Micheline Lesaffre, Marc Leman, Bernard De Baets, Hans De Meyer · 2003
In this paper a new auditory model-based transcriber of vocal melodic queries is presented. Our experiments show that the new system can transcribe queries with an accuracy between 76 % (whistling) and 85 % (singing with syllables), and that it outperforms four state-of-the-art systems it was compared with.