FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W16490680> ?p ?o ?g. }

• W16490680 abstract "While humans can separate a sound of interest amidst a cacophony of contending sounds in an echoic environment, machine-based methods lag behind in solving this task. This thesis thus aims at improving performance of audio separation algorithms when they are informed i.e. have access to source location information. These locations are assumed to be known a priori in this work, for example by video processing. Initially, a multi-microphone array based method combined with binarytime-frequency masking is proposed. A robust least squares frequency invariant data independent beamformer designed with the location information isutilized to estimate the sources. To further enhance the estimated sources, binary time-frequency masking based post-processing is used but cepstral domain smoothing is required to mitigate musical noise. To tackle the under-determined case and further improve separation performanceat higher reverberation times, a two-microphone based methodwhich is inspired by human auditory processing and generates soft time-frequency masks is described. In this approach interaural level difference,interaural phase difference and mixing vectors are probabilistically modeled in the time-frequency domain and the model parameters are learnedthrough the expectation-maximization (EM) algorithm. A direction vector is estimated for each source, using the location information, which is used asthe mean parameter of the mixing vector model. Soft time-frequency masks are used to reconstruct the sources. A spatial covariance model is then integrated into the probabilistic model framework that encodes the spatialcharacteristics of the enclosure and further improves the separation performancein challenging scenarios i.e. when sources are in close proximity andwhen the level of reverberation is high.Finally, new dereverberation based pre-processing is proposed based on the cascade of three dereverberation stages where each enhances the twomicrophonereverberant mixture. The dereverberation stages are based on amplitude spectral subtraction, where the late reverberation is estimated and suppressed. The combination of such dereverberation based pre-processing and use of soft mask separation yields the best separation performance. All methods are evaluated with real and synthetic mixtures formed for example from speech signals from the TIMIT database and measured room impulse responses." @default.
• W16490680 created "2016-06-24" @default.
• W16490680 creator A5038491744 @default.
• W16490680 date "2013-01-01" @default.
• W16490680 modified "2023-09-28" @default.
• W16490680 title "Informed algorithms for sound source separation in enclosed reverberant environments" @default.
• W16490680 cites W1495679096 @default.
• W16490680 cites W1500379785 @default.
• W16490680 cites W1514159701 @default.
• W16490680 cites W1540485935 @default.
• W16490680 cites W1546765626 @default.
• W16490680 cites W1572557551 @default.
• W16490680 cites W1651870529 @default.
• W16490680 cites W1832112969 @default.
• W16490680 cites W1964538581 @default.
• W16490680 cites W1969386733 @default.
• W16490680 cites W1981463705 @default.
• W16490680 cites W1989919048 @default.
• W16490680 cites W1991139021 @default.
• W16490680 cites W2023254078 @default.
• W16490680 cites W2025127228 @default.
• W16490680 cites W2031583051 @default.
• W16490680 cites W2034040413 @default.
• W16490680 cites W2043532615 @default.
• W16490680 cites W2057662515 @default.
• W16490680 cites W2057889776 @default.
• W16490680 cites W2060108923 @default.
• W16490680 cites W2062572859 @default.
• W16490680 cites W2063273764 @default.
• W16490680 cites W2068068201 @default.
• W16490680 cites W2088399049 @default.
• W16490680 cites W2091759147 @default.
• W16490680 cites W2096301647 @default.
• W16490680 cites W2096327012 @default.
• W16490680 cites W2098382639 @default.
• W16490680 cites W2098532729 @default.
• W16490680 cites W2098717567 @default.
• W16490680 cites W2102423079 @default.
• W16490680 cites W2103825654 @default.
• W16490680 cites W2104887782 @default.
• W16490680 cites W2104969945 @default.
• W16490680 cites W2107688514 @default.
• W16490680 cites W2110120390 @default.
• W16490680 cites W2110322414 @default.
• W16490680 cites W2114218864 @default.
• W16490680 cites W2116766578 @default.
• W16490680 cites W2117678320 @default.
• W16490680 cites W2120076267 @default.
• W16490680 cites W2120838354 @default.
• W16490680 cites W2122171611 @default.
• W16490680 cites W2122884709 @default.
• W16490680 cites W2127851351 @default.
• W16490680 cites W2128911505 @default.
• W16490680 cites W2135081730 @default.
• W16490680 cites W2135823751 @default.
• W16490680 cites W2136682440 @default.
• W16490680 cites W2137159737 @default.
• W16490680 cites W2141224535 @default.
• W16490680 cites W2142352693 @default.
• W16490680 cites W2143027228 @default.
• W16490680 cites W2143673560 @default.
• W16490680 cites W2144404214 @default.
• W16490680 cites W2146053319 @default.
• W16490680 cites W2147271331 @default.
• W16490680 cites W2148692373 @default.
• W16490680 cites W2149942017 @default.
• W16490680 cites W2155699610 @default.
• W16490680 cites W2158452199 @default.
• W16490680 cites W2161947961 @default.
• W16490680 cites W2166458192 @default.
• W16490680 cites W2168010845 @default.
• W16490680 cites W2170768669 @default.
• W16490680 cites W2188162373 @default.
• W16490680 cites W2296319761 @default.
• W16490680 cites W2403555780 @default.
• W16490680 cites W2489292218 @default.
• W16490680 cites W2539207221 @default.
• W16490680 cites W2911394281 @default.
• W16490680 cites W2955959362 @default.
• W16490680 cites W2970850616 @default.
• W16490680 cites W3127686677 @default.
• W16490680 cites W3139695047 @default.
• W16490680 cites W3205392798 @default.
• W16490680 cites W630361185 @default.
• W16490680 cites W793691323 @default.
• W16490680 cites W2283455280 @default.
• W16490680 hasPublicationYear "2013" @default.
• W16490680 type Work @default.
• W16490680 sameAs 16490680 @default.
• W16490680 citedByCount "0" @default.
• W16490680 crossrefType "dissertation" @default.
• W16490680 hasAuthorship W16490680A5038491744 @default.
• W16490680 hasConcept C11413529 @default.
• W16490680 hasConcept C121332964 @default.
• W16490680 hasConcept C154945302 @default.
• W16490680 hasConcept C169760540 @default.
• W16490680 hasConcept C19118579 @default.
• W16490680 hasConcept C24890656 @default.
• W16490680 hasConcept C26760741 @default.
• W16490680 hasConcept C2776864781 @default.

• next