数据整理以映射录音中音符的时间和频率

Question

I have a dataset having 216 sound recording files represented by the start and end timing (in seconds) and the respective frequencies of the notes sung by male and female birds:

note	sound.files	start	end	freq	sex
1	a	2.7	3.2	1.55	f
2	a	3.2	3.6	1.17	m
3	a	3.6	4.0	1.17	f
4	a	3.9	4.3	0.89	m
5	a	4.3	4.4	0.79	f
1	b	1.9	2.3	1.45	f
2	b	2.4	2.7	3.71	m
3	b	2.6	3.1	1.36	f
4	b	3.1	3.4	3.62	m
5	b	3.9	4.4	0.79	m
6	b	4.5	4.6	1.17	f

I require to transform the data in a long format with time-mapped frequency values of male and female birds for each recording, eg:

sound.files	time	m	f
a	2.7	0	1.55
a	2.8	0	1.55
a	2.9	0	1.55
a	3.0	0	1.55
a	3.1	0	1.55
a	3.2	1.17	0
a	3.3	1.17	0

I tried the following code but it did not work and running into an error:

Error: Problem with summarise() input ..1 . x object 'freq' not found:

code:

cum_call1 <- function(start,end,freq){
  data.frame(time = seq(start,end,by = .1), calling = 1, freq= mean(freq))
}

cum_expand1 <- function(start,end){
  data.frame(time = seq(start,end,by = .1))
}

data.frame$start <- round(data.frame$start,1)
data.frame$end <- round(data.frame$end,1)


duet_call <- data.frame %>% 
  group_by(sound.files,sex,note) %>% 
  summarise(cum_call1(start,end,freq)) %>% 
  ungroup() %>% 
  select(-note)

Is there any right/better way to go about it? Any suggestions are welcome! Thanks in advance!

Answer 1

Not base R but the pivot_wider function from one of the tidyverse packages should help. You're pivoting the column to be wider not longer as one column (sex) is becoming two.(m,f)

library(tidyverse)

pivot_wider(data,names_from=sex,values_from=freq)