Abstract | ||
---|---|---|
A monaural noise suppression algorithm is proposed based on filtering the spectro-temporal modulations of noisy speech. The modulations are estimated from a multiscale representation of the signal spectrograrn generated by a model of sound processing in the auditory system. A significant advantage of this method is its ability to suppress noise that has distinctive modulation patterns, despite being spectrally overlapping with the speech. The performance of the algorithm is evaluated using subjective and objective tests and compared to the Optimal Smoothing and Minimum Statistics approach (R. Martin 2001). The results demonstrate the efficacy of the spectro-temporal filtering approach in the conditions examined. |
Year | DOI | Venue |
---|---|---|
2005 | 10.1109/ICASSP.2005.1415311 | 2005 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS 1-5: SPEECH PROCESSING |
Keywords | Field | DocType |
filtering,acoustic noise,spectrogram,objective tests,performance,psychology,sound processing | Noise,Speech enhancement,Pattern recognition,Spectrogram,Computer science,Filter (signal processing),Speech recognition,Smoothing,Artificial intelligence,Audio signal processing,Monaural,Modulation (music) | Conference |
ISSN | Citations | PageRank |
1520-6149 | 7 | 0.74 |
References | Authors | |
7 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Nima Mesgarani | 1 | 256 | 22.43 |
Shihab Shamma | 2 | 554 | 67.25 |