Wed PM2.L1.3
HSP-TL: A DEEP METRIC LEARNING MODEL WITH TRIPLET LOSS FOR HIT SONG PREDICTION
Petros Vavaroutsos, Pantelis Vikatos, Orfium Research, Greece
Session:
Wed PM2.L1: Music signal analysis and processing Lecture
Track:
ASMSP - Acoustic, Speech and Music Signal Processing
Location:
EUROPAEA
Presentation Time:
Wed, 6 Sep, 16:50 - 17:10 Finland Time (UTC +3)
Session Chair:
Toon van Waterschoot, KU Leuven
Presentation
Discussion
Resources
No resources available.
Session Wed PM2.L1
Wed PM2.L1.1: DAACI-VODAN: IMPROVING VOCAL DETECTION WITH NEW DATA AND METHODS
Helena Cuesta, Daaci Ltd., United Kingdom; Nadine Kroher, Aggelos Pikrakis, Time Machine Capital 2 Ltd., United Kingdom; Stojan Djordjevic, Daaci Ltd., United Kingdom
Wed PM2.L1.2: On-line Chord Recognition Using FifthNet with Synchrosqueezing Transform
Rikuto Ito, Natsuki Akaishi, Waseda University, Japan; Kohei Yatabe, Tokyo University of Agriculture and Technology, Japan; Yasuhiro Oikawa, Waseda University, Japan
Wed PM2.L1.3: HSP-TL: A DEEP METRIC LEARNING MODEL WITH TRIPLET LOSS FOR HIT SONG PREDICTION
Petros Vavaroutsos, Pantelis Vikatos, Orfium Research, Greece
Wed PM2.L1.4: ONSETS AND VELOCITIES: AFFORDABLE REAL-TIME PIANO TRANSCRIPTION USING CONVOLUTIONAL NEURAL NETWORKS
Andres Fernandez, University of Tuebingen, Germany
Wed PM2.L1.5: HYBRID Y-NET ARCHITECTURE FOR SINGING VOICE SEPARATION
Rashen Fernando, Pamudu Ranasinghe, Udula Ranasinghe, Janaka Wijayakulasooriya, Pantaleon Perera, University of Peradeniya, Sri Lanka