dspp_metadataParameters: Provides the user with speaker dependent signal processing...

In humlab-speech/reindeer: This package extends the capabilities of emuR to work with speech data in a nordic climate

Provides the user with speaker dependent signal processing parameters

Description

This function returns age and gender specific speech signal processing parameters. Normally, the function just returns the stored /pre-computed DSPP tibble, which contains the output of this when applied previously.

Usage

dspp_metadataParameters(recompute = FALSE)

Arguments

recompute

If TRUE, this function will estimate age appropriate settings from the collection of empirical defaults stored in this package. If FALSE, this function will just return the stored DSPP data.

Details

Recomputation may be required if more data has been added to the file file.path(system.file(package = "reindeer",mustWork = TRUE),"default_parameters.xlsx"). If re-computation has been requested, this function then expands the age ranges in the spreadsheet file and computes a Gender stratified locally-weighted polynomial regression with age as the independent variable and the parameter value as the dependent variable. Separate models are computed for Male and Female speakers. Additionally, a separate model of all parameters are computed across all data for the case where the gender of a speaker is unknown (Gender==NA). Parameters that have previously been estimated using the function are are stored in the DSPP data frame.

Value

A tibble::tibble with one row per age and gender combination, and with one column per parameter for which an age dependent setting was identified in the set of parameters.

humlab-speech/reindeer documentation built on May 21, 2023, 4:43 p.m.