Norah's Phoneme Question

gistfile1.txt

## Norah's Question:  Are there any words that start with chr (phoenetically /k/ /r/) that don't have a a short i sound following it?

library(openssl)
library(textshape)
library(tidyverse)

cmudict <- readLines('https://raw.githubusercontent.com/michelleful/ToBoldlyStress/master/stressed_spelling.txt')
cmudict7b <- readLines('http://svn.code.sf.net/p/cmusphinx/code/trunk/cmudict/cmudict-0.7b') %>% tail(-121) %>% head(-4)


stress_key <- tribble(
~stress_code,  ~stress,
     0,  'No stress',
     1,  'Primary stress',
     2,  'Secondary stress'
)



phonemes <- tibble(
    raw = grep('\\(\\d+\\)', cmudict7b, value = TRUE, invert = TRUE)
) %>%
    dplyr::filter(grepl('^[A-Z]', raw)) %>%
    filter(grepl('^CH', raw)) %>%
    mutate(
        id = openssl::md5(as.character(seq_len(n()))),
        word = tolower(gsub('  .+$', '', raw)),
        phoneme = trimws(gsub('^.  ', '', raw)) %>% lapply(function(x) unlist(strsplit(x, '\\s+')))
    ) %>%
    select(-raw) %>%
    unnest(cols = c(phoneme)) %>%
    mutate(
        stress_code = as.integer(gsub('\\D', '', phoneme)),
        phoneme = gsub('\\d', '', phoneme)
    ) %>%
    left_join(stress_key, by = 'stress_code') 


ch_k_sample <- phonemes %>%
    group_by(word) %>%
    slice(2) %>%
    filter(phoneme == 'K')


ch_k_sample %>%
    filter(grepl('^chr[iy]', word)) %>%
    select(word) %>%
    print(n = Inf)

ch_k_sample %>%
    filter(!grepl('^chr[iy]', word))%>%
    select(word) %>%
    print(n = Inf)