In MarcelleO/sound_surfeR: Automatic Acoustic Template Matching

library(tuneR, warn.conflicts = F, quietly = T)
library(ggfortify) 
library(ggplot2)



#read in beep audio file 
#load("data/beep1.rda")
str(beep1)



# determine duration and frequency 
dur = length(beep1)/beep1@samp.rate
dur # seconds
fs = beep1@samp.rate
fs # Hz


# extract signal
snd = beep1@left
class(snd)



# demean to remove DC offset and clean audio 
#DC offset = mean amplitude displacement from zero

snd = snd - mean(snd)

#or using our function 
demean <- function(x){
  class(x) == "numeric"
  x <- x-mean(x)
  return(x)}

snd = demean(snd)

# plot waveform
plot(snd, type = 'l', xlab = 'Samples', ylab = 'Amplitude')


#The next step is to define the parameters used to construct the spectrogram. Here I am using values that are appropriate for our sound files. 

# number of points to use for the fft
nfft=1024
# window size (in points)
window=256
# overlap (in points)
overlap=128





# create spectrogram
library(signal, warn.conflicts = F, quietly = T) # signal processing functions

library(oce, warn.conflicts = F, quietly = T) # image plotting functions and nice color maps

spec = specgram(x = snd,
                n = nfft,
                Fs = fs,
                window = window,
                overlap = overlap
)
# discard phase information
P = abs(spec$S)
# normalize
P = P/max(P)
# convert to dB
P = 10*log10(P)
# config time axis
t = spec$t
# plot spectrogram
imagep(x = t,
       y = spec$f,
       z = t(P),
       col = oce.colorsViridis,
       ylab = 'Frequency [Hz]',
       xlab = 'Time [s]',
       drawPalette = T,
       decimate = F
)









# read in clap audio file
load("../data/clap1.rda")
clap1
# extract signal
snd = clap1@left

# determine duration and frequency 
dur = length(clap1)/clap1@samp.rate
dur # seconds
fs = beep1@samp.rate
fs # Hz

getwd()

# demean to remove DC offset and clean audio 
#DC offset = mean amplitude displacement from zero
snd = snd - mean(snd)

# plot waveform
plot(snd, type = 'l', xlab = 'Samples', ylab = 'Amplitude')


#The next step is to define the parameters used to construct the spectrogram. Here I am using values that are appropriate for our sound files. 

# number of points to use for the fft
nfft=1024
# window size (in points)
window=256
# overlap (in points)
overlap=128



# create spectrogram

spec = specgram(x = snd,
                n = nfft,
                Fs = fs,
                window = window,
                overlap = overlap
)


# discard phase information
P = abs(spec$S)
# normalize
P = P/max(P)
# convert to dB
P = 10*log10(P)
# config time axis
t = spec$t
# plot spectrogram
imagep(x = t,
       y = spec$f,
       z = t(P),
       col = oce.colorsViridis,
       ylab = 'Frequency [Hz]',
       xlab = 'Time [s]',
       drawPalette = T,
       decimate = F
)













# read in note audio file
load("data/note1.rda")

# extract signal
snd = note1@left

# determine duration and frequency 
dur = length(note1)/note1@samp.rate
dur # seconds
fs = beep1@samp.rate
fs # Hz

# demean to remove DC offset and clean audio 
#DC offset = mean amplitude displacement from zero
snd = snd - mean(snd)

# plot waveform
plot(snd, type = 'l', xlab = 'Samples', ylab = 'Amplitude')


#The next step is to define the parameters used to construct the spectrogram. Here I am using values that are appropriate for our sound files. 

# number of points to use for the fft
nfft=1024
# window size (in points)
window=256
# overlap (in points)
overlap=128





# create spectrogram

spec = specgram(x = snd,
                n = nfft,
                Fs = fs,
                window = window,
                overlap = overlap
)

# discard phase information
P = abs(spec$S)

# normalize
P = P/max(P)

# convert to dB
P = 10*log10(P)

# config time axis
t = spec$t

# plot spectrogram
imagep(x = t,
       y = spec$f,
       z = t(P),
       col = oce.colorsViridis,
       ylab = 'Frequency [Hz]',
       xlab = 'Time [s]',
       drawPalette = T,
       decimate = F
)












#now we can use monitoR to make our templates so that we can compare new unknown audio files to files with defined parameters and generate a signal score. 

library(monitoR)

#create file list path
fileList <- list.files(path = "data", full.names = TRUE)
fileList

data
getwd()
#create binary match point template 1 for beep sound 
Temp_beep1 <- makeBinTemplate("data/3beep1.wav", frq.lim = c(3, 6), t.lim = c(0, .7), name = "3beep1", amp.cutoff = (-25))

#template 2 for beep sound .
Temp_beep2 <- makeBinTemplate("data/3beep2.wav", frq.lim = c(3, 6), t.lim = c(0, .7), name = "3beep2", amp.cutoff = (-25))

#template 1 for clap sound 
Temp_clap1 <- makeBinTemplate("data/clap1.wav", frq.lim = c(0.1, 6), t.lim = c(0, .5), name = "clap1", amp.cutoff = (-25))

#template 2 for clap sound 
Temp_clap2 <- makeBinTemplate("data/clap2.wav", frq.lim = c(0.1, 6), t.lim = c(0, .5), name = "clap2", amp.cutoff = (-25))

#template 1 for note sound 
Temp_note1 <- makeBinTemplate("data/note1.wav", frq.lim = c(0.1, 6), t.lim = c(0, .5), name = "note1", amp.cutoff = (-25))

#template 2 for note sound 
Temp_note2 <- makeBinTemplate("data/note2.wav", frq.lim = c(0.1, 6), t.lim = c(0, .5), name = "note2", amp.cutoff = (-25))




#Create a bin with the templates you want to use for automatic matching. 
tempbin <- combineBinTemplates(Temp_beep1, Temp_beep2, Temp_clap1, Temp_clap2) 
str(tempbin)




#set file path for novel sound file you wish to compare to files within templates
load("data/beep3.rda")
str(beep3)

#use binMatch function to conduct binary matchpoint analysis   
bscores <- binMatch("data/3beep3.wav", tempbin, warn = FALSE)

bdetects <- findPeaks(bscores)
bdetects

getDetections(bdetects)






#lets try another novel sound file 
#set file path for novel sound file you wish to compare to files within templates
load("data/clap3.rda")
str(clap3)



#use binMatch function to conduct binary matchpoint analysis   
bscores <- binMatch("data/clap3.wav", tempbin, warn = FALSE)

bdetects <- findPeaks(bscores)
bdetects

getDetections(bdetects)


#alternatively you can use our new function 
bma <- function(wav_file, tempbin){
  class(wav_file) == "wave"
  class(tempbin) == "binTemplateList"
  #require monitoR
  bscores <- binMatch(wav_file, tempbin, warn = FALSE)
  bdetects <- findPeaks(bscores)
  scores <- getDetections(bdetects)
  return(scores)}

bma("data/clap3.wav", tempbin)




#create spectrograph cross correlation template 1 for beep sound 
Temp_beep1 <- makeCorTemplate("data/3beep2.wav", frq.lim = c(3, 6), t.lim = c(0, .7), name = "3beep1", amp.cutoff = (-25))

#template 2 for beep sound .
Temp_beep2 <- makeCorTemplate("data/3beep2.wav", frq.lim = c(3, 6), t.lim = c(0, .7), name = "3beep2", amp.cutoff = (-25))

#template 1 for clap sound 
Temp_clap1 <- makeCorTemplate("data/clap1.wav", frq.lim = c(0.1, 6), t.lim = c(0, .5), name = "clap1", amp.cutoff = (-25))

#template 2 for clap sound 
Temp_clap2 <- makeCorTemplate("data/clap2.wav", frq.lim = c(0.1, 6), t.lim = c(0, .5), name = "clap2", amp.cutoff = (-25))

#template 1 for note sound 
Temp_note1 <- makeCorTemplate("data/note1.wav", frq.lim = c(0.1, 6), t.lim = c(0, .5), name = "note1", amp.cutoff = (-25))

#template 2 for note sound 
Temp_note2 <- makeCorTemplate("data/note2.wav", frq.lim = c(0.1, 6), t.lim = c(0, .5), name = "note2", amp.cutoff = (-25))


tempcor <- combineCorTemplates(Temp_beep1, Temp_beep2, Temp_clap1, Temp_clap2)


#set file path for novel sound file you wish to compare to files within templates
load("data/beep3.rda")
str(beep3)

#use corMatch function to conduct binary matchpoint analysis   
cscores <- corMatch("data/3beep3.wav", tempcor, warn = FALSE)

cdetects <- findPeaks(cscores)
cdetects

getDetections(cdetects)


#alternatively you can use our new function 
cma <- function(wav_file, tempcor){
  class(wav_file) == "wave"
  class(tempcor) == "CorTemplateList"
  #require monitoR
  cscores <- corMatch(wav_file, tempcor, warn = FALSE)
  cdetects <- findPeaks(cscores)
  scores <- getDetections(cdetects)
  return(scores)}

cma("data/clap3.wav", tempcor)