agricolamz

# This code is licensed under the terms of the MIT license
# Author: George Moroz
# Date: 2025-08-14
# Update: 2025-08-27
# see the oficial docs: https://ruscorpora.github.io/public-api/

library(tidyverse)
my_token <- "put your token here"
lemma_for_search <- "накачать"

agricolamz

library(tidyverse)

readxl::read_xlsx("data.xlsx") |> 
  filter(corpus == "Dagestan.xml",
         !is.na(string_id)) ->
  df

df |> 
  select(where(function(x) sum(is.na(x)) != nrow(df))) |> 
  select(-person_id, -corpus) |> 

agricolamz

agricolamz

library(tidyverse)
read_csv("https://raw.githubusercontent.com/agricolamz/2024_HSE_b_da4l/master/data/Coretta_2017_icelandic.csv") |> 
  filter(speaker == "tt01") ->
  vowels

mean_data <- mean(vowels$vowel.dur)
sd_data <- sd(vowels$vowel.dur)

m1 <- function(x) dnorm(x, mean = mean_data, sd = sd_data) * 
  dnorm(x, mean = 87, sd = 25)

agricolamz

library(tidyverse)
read_csv("https://raw.githubusercontent.com/agricolamz/2025_HSE_b_da4l/refs/heads/main/data/Coretta_2017_icelandic.csv") |> 
  filter(speaker == "tt01") ->
  vowels

sd_prior <- 25
sd_data <- sd(vowels$vowel.dur)
sd_post <- 1/sqrt(1/sd_prior^2 + 1/sd_data^2)
mean_prior <- 87
mean_data <- mean(vowels$vowel.dur)

agricolamz

library(tidyverse)
library(lingtypology)
df <- bivaltyp.feature()


df |> 
  filter(family_WALS == "Nakh-Daghestanian") |> 
  writexl::write_xlsx("~/Desktop/daghestan_bivaltyp.xlsx")

df |> 

agricolamz

library(tidyverse)
phonfieldwork::read_from_folder("...") |> 
  filter(tier == 2,
         content != "") |> 
  mutate(dur = time_end-time_start) |> 
  summarise(duration_minutes = sum(dur)/60)

agricolamz

library(tidyverse)
khi <- read_csv("russian_spoken_corpora_analysis/dialect_khislavichi_udpiped_mystemed.csv")
spi <- read_csv("russian_spoken_corpora_analysis/dialect_spiridonova_buda_udpiped_mystemed.csv")

khi |> 
  bind_rows(spi) |> 
  filter(mystem_pos == "S",
         str_detect(mystem_feats, "nom.pl"),
         str_detect(mystem_feats, "^[mn],")) |> 
  writexl::write_xlsx("~/Desktop/4Sveta_N_mn_nompl.xlsx")

agricolamz

library(tidyverse)

read_csv("andic_russian_bor2.csv")  |> 
  mutate(language_material = str_count(ipa_comparison, "-"),
         russian_material = str_count(russian_ipa_comparison, "-"),
         diff = if_else(russian_material == language_material, "", "problem")) |> 
  filter(diff != "problem") ->
  df

df |> 

agricolamz

library(tidyverse)

input <- "Мой друг, которого я давно не видел, крепко спит."

tmp_input <- tempfile(fileext = ".txt")
tmp_output <- tempfile(fileext = ".txt")
write_lines(input, tmp_input)

str_c("mystem -cdni --eng-gr ", tmp_input, " ", tmp_output) |> 
  system()