SpeeD @ MediaEval 2014: Spoken Term Detection with Robust Multilingual Phone Recognition
Andi Buzo, Horia Cucu, Corneliu Burileanu · 2014
In this paper, we attempt to resolve the Spoken Term Detection (STD) problem for under-resourced languages by phone recognition with a multilingual acoustic model of three languages (Albanian, English and Romanian). The Power Normalized Cepstral Coefficients (PNCC) features are used for improved robustness to noise.