de.sciss.strugatzki.CrossSimilarity
The file to which the cross similarity vector is written as an audio file.
The file to which the cross similarity vector is written as an audio file.
The format type for the output file.
The format type for the output file.
The database folder is merely used to retrieve the normalization file,
given that normalize is true.
The database folder is merely used to retrieve the normalization file,
given that normalize is true.
Maximum energy boost (as an amplitude factor) allowed for a match to be considered.
Maximum energy boost (as an amplitude factor) allowed for a match to be considered.
The estimation of the boost factor for two matched signals
is exp ((ln( loud_in ) - ln( loud_db )) / 0.6 )
The XML file holding the extractor parameters corresponding to the first audio input file.
The XML file holding the extractor parameters corresponding to the first audio input file. The audio input file's feature vector output file is determined from this meta file.
The XML file holding the extractor parameters corresponding to the second audio input file.
The XML file holding the extractor parameters corresponding to the second audio input file. The audio input file's feature vector output file is determined from this meta file.
Whether to apply normalization to the features (recommended).
Whether to apply normalization to the features (recommended).
An option which restricts the calculation to a given span within the first input file.
An option which restricts the calculation to a given span within the
first input file. If Span.all, the whole file is considered.
An option which restricts the calculation to a given span within the second input file.
An option which restricts the calculation to a given span within the
second input file. If Span.all, the whole file is considered.
The balance between the feature of loudness curve and spectral composition (MFCC).
The balance between the feature of loudness curve and spectral composition (MFCC). A value of 0.0 means the segmentation is only performed by considering the spectral features, and a value of 1.0 means the segmentation is taking only the loudness into consideration. Values in between give a measure that takes both features into account with the given priorities.