如何正确将音素转写字符串分割为单个音素段？

你遇到的核心问题是正则表达式的匹配顺序：你的phonemes向量里，单字符的短元音i排在双字符的长元音iː前面，导致正则引擎优先匹配了短的i，而不是完整的iː，最终把长元音拆成了两个部分。

解决方案：调整音素匹配顺序 #

正则的交替匹配是从左到右尝试的，所以我们只需要把多字符的音素（双元音、长元音、塞擦音）都放在单字符音素的前面，让引擎优先匹配完整的长音素，而不是先匹配它的子字符。

步骤如下：

1. 重新排序你的音素向量：

# 优先排列多字符音素，再放单字符音素
phonemes <- c(
  # 双元音（双字符）
  "ɪə","eɪ","ʊə","ɔɪ","aɪ","eə","aʊ","əʊ",
  # 长元音（双字符）
  "iː","uː","ɜː","ɔː","ɑː",
  # 塞擦音（双字符）
  "ʧ","ʤ",
  # 单字符音素
  "ə","ɪ", "ɛ","ɒ","ʌ","æ","i","ʊ",
  "j","w",
  "r","l",
  "n","m","ŋ",
  "f","v","θ","ð","s","z","ʃ","ʒ","h",
  "p","b","t","d","k","g"
)

2. 重新生成匹配模式并执行分割：

library(stringr)

test1 <- "dɪd ɛnɪbɒdi liːv ðeə glɑːsɪz hɪə lɑːst wiːk sʌmbədi dɪd"
phonemes_pattern <- paste0("(", paste0(phonemes, collapse = "|"), ")")
test1_clean <- gsub(" ", "", test1)

# 执行分割
split_result <- str_split(test1_clean, paste0("(?<=", phonemes_pattern, ")"))

# 查看结果
split_result[[1]]

正确结果示例 #

调整后，长元音iː会被作为一个整体保留，分割结果里的对应部分会是：
"l", "iː", "v" 而不是之前的 "l", "i", "ː", "v"

整个分割结果会变成：

[1] "d"   "ɪ"   "d"   "ɛ"   "n"   "ɪ"   "b"   "ɒ"   "d"   "i"   "l"   "iː"  "v"   "ð"   "eə"  "g"   "l"   "ɑː"  "s"   "ɪ"   "z"   "h"   "ɪ"   "ə"  
[25] "l"   "ɑː"  "s"   "t"   "w"   "iː"  "k"   "s"   "ʌ"   "m"   "b"   "ə"   "d"   "i"   "d"   "ɪ"   "d"   ""

这样就能完美解决长元音被错误拆分的问题啦！

内容的提问来源于stack exchange，提问作者Chris Ruehlemann