Work

Compact long context spectral factorisation models for noise robust recognition of medium vocabulary speech

Year: 2013
Type: conference-paper
Abstract: In environments containing multiple non-stationary sound sources, it becomes increasingly difficult to recognise speech from its shorttime spectra alone. Long-context speech and noise models, where phonetic patterns and noise events may span hundreds of milliseconds, have been found beneficial in such separation tasks. Thus far the majority of work employing non-negative matrix factorisation to long-context spectrogram separation has been conducted on small vocabulary tasks by exploiting large speech and noise dictionaries containing thousands of atoms. In this work we study whether the previously proposed factorisation methods are applicable to more natural speech and limited noise context while keeping the model sizes practically feasible. Results are evaluated on the WSJ0 5k-based 2nd CHiME Challenge Track 2 corpus, where we achieve approximately 4 % absolute improvement in speech recognition rates compared to baseline using the proposed enhancement framework. Index Terms — Spectral factorisation, speech recognition, noise robustness 1.
Source: Lirias
Language: English
Open Access status: green
OpenAlex ID: W2288838913
FWCI: 1.869
Cites: 18
Cited by: 8
Related to: 10