Ram C. M. C. Shekar — accepted AI-conference papers

2026

LIWHIZ: A NON-INTRUSIVE LYRIC INTELLIGIBILITY PREDICTION SYSTEM FOR THE CADENZA CHALLENGE

ICASSP 2026poster

We present LIWhiz, a non-intrusive lyric intelligibility prediction system submitted to the ICASSP 2026 Cadenza Challenge. LIWhiz leverages Whisper for robust feature extraction and a trainable back-end for score prediction. Tested on the Cadenza Lyric Intelligibility Prediction (CLIP) evaluation se…

Cited by 0 Source ↗PDF ↗Scholar ↗

2023

Filterbank Learning for Noise-Robust Small-Footprint Keyword Spotting

Iván López-Espejo, Ram C. M. C. Shekar, Zheng-Hua Tan, Jesper Jensen, John H. L. Hansen

ICASSP 2023accepted

In the context of keyword spotting (KWS), the replacement of handcrafted speech features by learnable features has not yielded superior KWS performance. In this study, we demonstrate that filterbank learning outperforms handcrafted speech features for KWS whenever the number of filterbank channels i…

Cited by 0 Source ↗Scholar ↗