normalizeData: Normalize audio data using Box-Cox transformation

In voiceR: Voice Analytics for Social Scientists

Normalize audio data using Box-Cox transformation

Description

This function normalizes audio data using the Box-Cox transformation. It takes in a data frame of audio data and a vector of measures to be normalized. Users can choose to normalize by dimensions and/or conditions.

Usage

normalizeData(
  audioData,
  measures = c("duration", "voice_breaks_percent", "RMS_env", "mean_loudness", "mean_F0",
    "sd_F0", "mean_entropy", "mean_HNR"),
  includeDimensions = FALSE,
  includeConditions = FALSE
)

Arguments

audioData	A data.frame generated by the autoExtract() function.
measures	A vector of strings specifying the measures to be normalized. Default corresponds to all the measures extracted by autoExtract().
includeDimensions	A logical value indicating whether or not to include dimensions in the normalization process. Default corresponds to FALSE.
includeConditions	A logical value indicating whether or not to include conditions in the normalization process. Default corresponds to FALSE.

Value

A list containing three elements: (1) a data frame of the normalized audio data and (2) a logical vector indicating whether or not each measure was transformed using Box-Cox transformation and (3) the Box-cox constant added to each measure.

Examples

normalizeData(testAudioData)

voiceR documentation built on Sept. 13, 2023, 1:07 a.m.