Abstract | ||
---|---|---|
We present a monaural approach to speech segregation that estimates the ideal binary mask (IBM) by combining amplitude modulation spectrogram (AMS) features, pitch-based features and speech presence probability (SPP) features derived from noise statistics. To maintain a high mask estimation accuracy in the presence of various background noises, the system employs environment-specific segregation models and automatically selects the appropriate model for a given input signal. Furthermore, instead of classifying each time-frequency (T-F) unit independently, the a posteriori probabilities of speech and noise presence are evaluated by considering adjacent T-F units. The proposed system achieves high classification accuracy. |
Year | DOI | Venue |
---|---|---|
2013 | 10.1109/WASPAA.2013.6701821 | Applications of Signal Processing to Audio and Acoustics |
Keywords | Field | DocType |
probability,signal classification,speech processing,time-frequency analysis,AMS,IBM,SPP,amplitude modulation spectrogram features,environment-aware ideal binary mask estimation,monaural approach,pitch-based features,speech presence probability features,speech segregation,time-frequency unit,background noise classification,ideal binary mask estimation,speech segregation | Speech processing,Noise statistics,Computer science,A priori and a posteriori,Artificial intelligence,Binary number,Pattern recognition,Spectrogram,Speech recognition,Amplitude modulation,Time–frequency analysis,Acoustics,Monaural | Conference |
ISSN | Citations | PageRank |
1931-1168 | 3 | 0.47 |
References | Authors | |
7 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Tobias May | 1 | 43 | 4.97 |
Torsten Dau | 2 | 56 | 10.01 |