Generic Normalization Procedure — norm

This is a generic normalization procedure with which you can create your own normalization method.

Usage

norm_generic(
  .data,
  ...,
  .by = NULL,
  .by_formant = FALSE,
  .by_token = FALSE,
  .L = 0,
  .S = 1,
  .pre_trans = function(x) x,
  .post_trans = function(x) x,
  .drop_orig = FALSE,
  .keep_params = FALSE,
  .names = "{.formant}_n",
  .silent = FALSE,
  .call = caller_env()
)

Arguments

.data: A data frame containing vowel formant data
...: <tidy-select> One or more unquoted expressions separated by commas. These should target the vowel formant data columns.
.by: <tidy-select> A selection of columns to group by. Typically a column of speaker IDs.
.by_formant: Whether or not the normalization method is formant intrinsic.
.by_token: Whether or not the normalization method is vowel intrinsic
.L: An expression defining the location parameter. See Details for more information.
.S: An expression defining the scale parameter. See Details for more information.
.pre_trans: A function to apply to formant values before normalization.
.post_trans: A function to apply to formant values after normalization.
.drop_orig: Whether or not to drop the original formant data columns.
.keep_params: Whether or not to keep the Location (*_.L) and Scale (*_.S) normalization parameters
.names: A glue::glue() expression for naming the normalized data columns. The "{.formant}" portion corresponds to the name of the original formant columns.
.silent: Whether or not the informational message should be printed.
.call: Used for internal purposes.

Value

A data frame of normalized formant values

Details

The following norm_* procedures are built on top of norm_generic().

Location and Scale expressions

All normalization procedures built on norm_generic produce normalized formant values ($\hat{F}$) by subtracting a location parameter ($L$) and dividing by a scale parameter ($S$).

$$ \hat{F} = \frac{F-L}{S} $$

The expressions for calculating $L$ and $S$ can be passed to .L and .S, respectively. Available values for these expressions are

.formant: The original formant value
.formant_num: The number of the formant. (e.g. 1 for F1, 2 for F2 etc)

Along with any data columns from your original data.

Pre and Post normalization transforms

To apply any transformations before or after normalization, you can pass a function to .pre_trans and .post_trans.

Formant In/Extrinsic Normalization

If .by_formant is TRUE, normalization will be formant intrinsic. If .by_formant is FALSE, normalization will be formant extrinsic.

Token In/Extrinsic Normalization

If .by_token is TRUE, normalization will be token intrinsic. If .by_token is FALSE, normalization will be token extrinsic.

Examples

library(tidynorm)
library(dplyr)

speaker_data |>
  norm_generic(
    F1:F3,
    .by = speaker,
    .by_formant = TRUE,
    .L = median(.formant, na.rm = TRUE),
    .S = mad(.formant, na.rm = TRUE),
    .drop_orig = TRUE,
    .names = "{.formant}_mad"
  )
#> Normalization info
#> • normalized `F1`, `F2`, and `F3`
#> • normalized values in `F1_mad`, `F2_mad`, and `F3_mad`
#> • grouped by `speaker`
#> • formant intrinsic
#> 
#> # A tibble: 10,697 × 9
#>      .id speaker vowel plt_vclass ipa_vclass  F1_mad  F2_mad F3_mad word    
#>    <int> <chr>   <chr> <chr>      <chr>        <dbl>   <dbl>  <dbl> <chr>   
#>  1     1 s01     EY    eyF        ejF         1.13    0.400  -0.521 OKAY    
#>  2     2 s01     AH    uh         ʌ           0.643  -0.166   0.629 UM      
#>  3     3 s01     AY    ay         aj          2.09   -0.0211  0.165 I'M     
#>  4     4 s01     IH    i          ɪ          -0.454  -1.13    1.41  LIVED   
#>  5     5 s01     IH    i          ɪ          -0.0221  1.04    1.03  IN      
#>  6     6 s01     AH    @          ə          -0.0206 -0.104   0.715 COLUMBUS
#>  7     7 s01     AY    ay         aj         -0.280   0.0479  0.968 MY      
#>  8     8 s01     IH    i          ɪ          -0.649   1.07    1.03  ENTIRE  
#>  9     9 s01     ER    *hr        ə˞         -0.585  -0.709  -2.35  ENTIRE  
#> 10    10 s01     AY    ay0        aj0         0.987  -0.656  -2.34  LIFE    
#> # ℹ 10,687 more rows