FAVE Remeasure module¶
- class fave.extract.remeasure.VowelMeasurement¶
represents a vowel measurement
- fave.extract.remeasure.calculateVowelMeans(vowels)¶
calculates [means] and [covariance matrices] for each vowel class. It returns these as numpy arrays in dictionaries indexed by the vowel class.
- fave.extract.remeasure.createVowelDictionary(measurements)¶
Creates a dictionary of F1, F2, B1, B3 and Duration observations by vowel type. vowel index indicates the index in lines[x] which should be taken as identifying vowel categories.
- fave.extract.remeasure.excludeOutliers(vowels, vowelMeans, vowelCovs)¶
Finds outliers and excludes them.
- fave.extract.remeasure.loadfile(file)¶
Loads an extractFormants file. Returns formatted list of lists.
- fave.extract.remeasure.output(remeasurements)¶
writes measurements to file according to selected output format
- fave.extract.remeasure.pruneVowels(vowels, vowel, vowelMeans, vowelCovs, outlie)¶
Tries to prune outlier vowels, making sure enough tokens are left to calculate mahalanobis distance.
- fave.extract.remeasure.repredictF1F2(measurements, vowelMeans, vowelCovs, vowels)¶
Predicts F1 and F2 from the speaker’s own vowel distributions based on the mahalanobis distance.