If just one value is provided, that number will be used for all intervals. Each interval can have a different number. Samples: number of samples taken in each surveyed interval.The value on the right field has to be equal or greater than the value on the left field. Interval range: indices of the first and last intervals in the specified tier to be sampled.Transformation: options to transform the F0 values from the Hz scale to semitones relative to different reference levels, the OctaveMedian scale or z-score normalization.The default value is NA, which is the string used by R to represent missing values. Unvoiced: the user can select what string will be used when the script samples an unvoiced part of the F0 contour.Options are quadratic, linear or no interpolation. Interpolation: Which kind of interpolation apply in voiceless intervals.Bandwidth: how much smoothing should be applied (the greater the number, the smoother the contour).Smooth: allows users to choose whether or not smoothing should be applied to F0 contours.Tier: number of the TextGrid tier to be analyzed.

Report: Path of the folder and name of the file (with extension) of the report outputted by the script.They can be the same or different folders. Pitch folder and Grid folder: path of folders where Pitch and TextGrid files are stored.For instance, if the user only wants to analyze files whose name start with the string subj1, the appropriate pattern would be subj1*. If the user wants to select only a subset of the files, a pattern can be specified. The default option is *, which causes the script to work on all the files on the selected folders. Pattern: Wildcard characters * and ? can be used to specify a pattern that all file names must have in order to be selected.Upon running the script, a window like the one below will appear, where the user has to set a number of parameters. Tab-separated text file containing time-normalized F0 values and metadata for each F0 value: file name, interval number with reference to non-empty intervals, sample number, normalization time step for each interval and timestamp of sampled points. One or more Pitch files and corresponding TextGrids with at least one interval tier containing non-empty intervals. Usually an interval will span a segment, a syllable or even whole words depending on the problem at hand. Time normalization is achieved by taking a user-defined number of evenly spaced F0 values for each interval defined in a TextGrid.

The script will generate a normalized F0 contour from a raw F0 curve. Praat script to perform time-normalization of F0 contours.

Mean f0 in praat