WE2.PA2.2

POST-TRAINING LATENT DIMENSION REDUCTION IN NEURAL AUDIO CODING

Thomas Muller, Stéphane Ragot, Pierrick Philippe, Orange Innovation, France; Pascal Scalart, IRISA - University of Rennes, France

Session:
WE2.PA2: Machine Learning for Audio and Acoustics Poster

Track:
ASMSP - Acoustic, Speech and Music Signal Processing

Location:
Poster Area 2

Presentation Time:
Wed, 28 Aug, 14:00 - 15:40 France Time (UTC +1)

Session Co-Chairs:
Rainer Martin, Ruhr-Universität Bochum and Nobutaka Ono, Tokyo Metropolitan University
Presentation
Discussion
Resources
No resources available.
Session WE2.PA2
WE2.PA2.1: DIVERSITY-BASED SAMPLING FOR IMBALANCED DOMAIN ADAPTATION
Andrea Napoli, Paul White, University of Southampton, United Kingdom
WE2.PA2.2: POST-TRAINING LATENT DIMENSION REDUCTION IN NEURAL AUDIO CODING
Thomas Muller, Stéphane Ragot, Pierrick Philippe, Orange Innovation, France; Pascal Scalart, IRISA - University of Rennes, France
WE2.PA2.3: Improving Speech Inversion Through Self-Supervised Embeddings and Enhanced Tract Variables
Ahmed Adel Attia, Yashish Siriwardena, Carol Espy-Wilson, University of Maryland, United States
WE2.PA2.4: USING RANDOM CODEBOOKS FOR AUDIO NEURAL AUTOENCODERS
Benoît Giniès, Xiaoyu Bie, Olivier Fercoq, Gaël Richard, Télécom Paris, Institut Polytechnique de Paris, France; ,
WE2.PA2.5: Interpreting End-to-End Deep Learning Models for Speech Source Localization Using Layer-wise Relevance Propagation
Luca Comanducci, Fabio Antonacci, Augusto Sarti, Politecnico di Milano, Italy
WE2.PA2.6: Fusing Audio and Metadata Embeddings Improves Language-based Audio Retrieval
Paul Primus, Gerhard Widmer, Johannes Kepler University, Austria
WE2.PA2.7: WEIGHT LIGHT, HEAR RIGHT: HEART SOUND CLASSIFICATION WITH A LOW-COMPLEXITY MODEL
Jiahao Ji, Lixian Zhu, Haojie Zhang, Kun Qian, Beijing Institute of Technology, China; Kele Xu, National University of Defense Technology, China; Zikai Song, Bin Hu, Beijing Institute of Technology, China; Björn W. Schuller, Technical University of Munich, Germany; Yoshiharu Yamamoto, The University of Tokyo, Japan
WE2.PA2.8: Audio-based Step-count Estimation for Running – Windowing and Neural Network Baselines
Philipp Wagner, Andreas Triantafyllopoulos, Alexander Gebhard, Björn Schuller, Technical University of Munich, Germany
WE2.PA2.9: Towards Green AI : Assessing the Robustness of Conformer and Transformer Models under Compression
Leila Ben Letaifa, LINEACT - CESI, France; Jean-Luc Rouas, LaBRI, France
WE2.PA2.10: DEEP DIGITAL JOINT SOURCE-CHANNEL BASED WIRELESS SPEECH TRANSMISSION
Mohammad Bokaei, Jesper Jensen, Aalborg university, Denmark; Simon Doclo, Oldenburg University, Germany; Jan Østergaard, Aalborg university, Denmark