FILE 01
active
Emotion from the raw waveform
CNN-n-GRU and related speech-analysis studies connect local acoustic patterns with temporal dependencies.
Open dossierSignal Press / editorial
Projects
Research dossiers linking methods, published studies and software.
The working files
FILE 02
active
Adaptive time-frequency learning
A research file connecting learned wavelets, wavelet packets, denoising and fractional superlets.
Open dossierFILE 03
active
Explainable feature boosting
Using feature attribution to refine acoustic representations and examine model decisions.
Open dossierFILE 04
active
Cross-attentive emotion fusion
Studying the relationship between audio, text and video through cross-attention and feature fusion.
Open dossierFrom a question to an implementation
Methods — The stages of the approaches and the choices behind them.
Publications — Full abstracts, author lists, citations and publisher links.
Software — The corresponding research code and reproducibility documentation.