Assign musical notes

Description

Assign musical notes in Scientific Pitch Notation or other variant. See voice::notes(). The notes are cut considering f0 to ensure alignment.

Usage

assign_notes(
  x,
  fmt = 0,
  min_points = 4,
  min_percentile = 0.75,
  max_na_prop = 1
)

Arguments

Examples

library(voice)
# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern = glob2rx('*.wav'), full.names = TRUE)
# Media dataset
M <- extract_features(path2wav)
assign_notes(M, fmt = 0) # f0
assign_notes(M, fmt = 1) # f1
assign_notes(M, fmt = 2) # f2

Returns the total time of audio files in seconds

Description

Returns the total time of audio files in seconds

Usage

audio_time(x, filesRange = NULL, recursive = FALSE)

Arguments

Value

A tibble containing file name <chr> and audio time <dbl> in seconds.

Examples

library(voice)

# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern <- glob2rx('*.wav'), full.names = TRUE)

# Tibble containing file name and audio time
(at <- voice::audio_time(unique(dirname(path2wav))))
str(at)

Check chords

Description

Check the sequence of musical notes for chords.

Usage

check_chords(x, window = 3, try_perm = FALSE)

Arguments

Examples

## Not run: 
library(voice)
check_chords(c('C','E','G'), window = 3, try_perm = FALSE)
check_chords(c('C','E','G'), window = 3, try_perm = TRUE)
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern = glob2rx('*.wav'), full.names = TRUE)
M <- extract_features(path2wav)
M$gain[is.na(M$f0)] <- NA
# assigning notes
f0_spn <- assign_notes(M, fmt = 0)
check_chords(f0_spn, window = 3, try_perm = FALSE)
check_chords(f0_spn, window = 3, try_perm = TRUE)
check_chords(f0_spn, window = 4, try_perm = TRUE)

## End(Not run)

Cut audio vectors

Description

Cut vectors

Usage

cut_audio(x, byvar = x)

Arguments

Examples

library(voice)
# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern = glob2rx('*.wav'), full.names = TRUE)
# Media dataset
M <- extract_features(path2wav)
cut_audio(M$f0)
cut_audio(M$gain, M$f0)

Who spoke when?

Description

Diarization of WAV audios.

Usage

diarize(
  fromWav,
  toRttm = NULL,
  autoDir = FALSE,
  pycall = "~/miniconda3/envs/pyvoice/bin/python",
  token = NULL
)

Arguments

Details

When autoDir = TRUE, the following directories are created: '../mp3','../rttm', '../split' and '../musicxml'. Use getwd() to find the parent directory '../'.

Value

RTTM files in NIST standard. See 'voice::read_rttm'.

Examples

## Not run: 
library(voice)

wavDir <- list.files(system.file('extdata', package = 'wrassp'),
pattern = glob2rx('*.wav'), full.names = TRUE)

voice::diarize(fromWav = unique(dirname(wavDir)),
toRttm = tempdir(),
token = NULL) # Must enter a token! See documentation.

(rttm <- dir(tempdir(), '.[Rr][Tt][Tt][Mm]$', full.names = TRUE))
file.info(rttm)

## End(Not run)

Duration of sequences

Description

Duration of sequences

Usage

duration(x, windowShift = 5)

Arguments

Value

A data frame with duration in number of lines/ocurrences (dur_line), milliseconds (dur_ms) and proportional (dur_prop).

Examples

library(voice)
duration(letters)
duration(c('a','a','a',letters,'z'))

nts <- c('NA','NA','A3','A3','A3','A3','A#3','B3','B3','C4','C4','C4','C4',
'C4','C4','C#4','C4','C4','C4','B3','A#3','NA','NA','NA','NA','NA','NA','NA',
'NA','NA','NA','NA','NA','NA','NA','NA','NA','NA','NA','D4','D4','D4','C#4',
'C#4','C#4','C4','C4','B3','B3','A#3','A#3','A3','A3','G3','G#3','G3','F#3')
duration(nts)

These functions are sourced from the 'embedr' package by Michael McCarthy, under MIT License: https://github.com/mccarthy-m-g/embedr/blob/master/LICENSE.md This inclusion is temporary and will be discontinued once 'embedr' is available on CRAN. See https://github.com/mccarthy-m-g/embedr for more details.

Description

Embed audio in R Markdown documents

Usage

embed_audio(
  src,
  type = c("mpeg", "ogg", "wav"),
  attribute = c("controls", "autoplay", "loop", "muted", "preload", "none"),
  id = "",
  placeholder = ""
)

Arguments

Details

'embed_audio()' provides a standard way to embed audio in R Markdown documents when the output format is HTML, and to print placeholder text when the output format is not HTML.

'embed_audio()' is a wrapper for the HTML5 '<audio>' element that prints HTML '<audio>' code in HTML documents built by R Markdown and placeholder text in non-HTML documents built by R Markdown. This function may be useful for conditional output that depends on the output format. For example, you may embed audio in an R Markdown document when the output format is HTML, and print placeholder text when the output format is LaTeX.

The function determines output format using [knitr::is_html_output()]. By default, these formats are considered as HTML formats: ‘c(’markdown', 'epub', 'html', 'html5', 'revealjs', 's5', 'slideous', 'slidy')'.

Value

If 'knitr::is_html_output()' is 'TRUE', returns HTML '<audio>' code. If 'knitr::is_html_output()' is 'FALSE', returns placeholder text.

Note

This function is supposed to be used in R code chunks or inline R code expressions. You are recommended to use forward slashes (/) as path separators instead of backslashes in the file paths.

Examples

# By default, embed_audio() embeds an audio element with playback controls
embed_audio(mp3)

# To change the attributes of the audio element, use `attribute`
embed_audio(mp3, attribute = c("controls", "loop"))

# To add placeholder text for non-HTML documents, use `placeholder`
embed_audio(mp3, placeholder = "This is placeholder text.")

## Not run: 
# embed_audio() is intended to be used in R Markdown code chunks or inline
# expressions. The following creates and knits an R Markdown document to
# HTML and PDF in your current working directory for you to inspect:
library(rmarkdown)
writeLines(c("# Hello embedr!",
"```{r embed-audio, echo=TRUE}",
"embed_audio(mp3, placeholder = 'This is placeholder text.')",
"```"), "test.Rmd")
render("test.Rmd", output_format = c('html_document', 'pdf_document'))

# Delete test files created by example code
unlink(c("test.Rmd", "test.html", "test.pdf"))

## End(Not run)

Embed video in R Markdown documents

Description

'embed_video()' provides a standard way to embed video in R Markdown documents when the output format is HTML, and to print placeholder text when the output format is not HTML.

Usage

embed_video(
  src,
  type = c("mp4", "webm", "ogg"),
  width = "320",
  height = "240",
  attribute = c("controls", "autoplay", "loop", "muted", "preload", "none"),
  thumbnail = NULL,
  id = "",
  placeholder = ""
)

Arguments

Details

'embed_video()' is a wrapper for the HTML5 '<video>' element that prints HTML '<video>' code in HTML documents built by R Markdown and placeholder text in non-HTML documents built by R Markdown. This function may be useful for conditional output that depends on the output format. For example, you may embed video in an R Markdown document when the output format is HTML, and print placeholder text when the output format is LaTeX.

The function determines output format using [knitr::is_html_output()]. By default, these formats are considered as HTML formats: ‘c(’markdown', 'epub', 'html', 'html5', 'revealjs', 's5', 'slideous', 'slidy')'.

Value

If 'knitr::is_html_output()' is 'TRUE', returns HTML '<video>' code. If 'knitr::is_html_output()' is 'FALSE', returns placeholder text.

Note

This function is supposed to be used in R code chunks or inline R code expressions. You are recommended to use forward slashes (/) as path separators instead of backslashes in the file paths.

Examples

# By default, embed_video() embeds a video element with playback controls
embed_video(mp4)

# To change the attributes of the video element, use `attribute`
embed_video(mp4, attribute = c("controls", "loop"))

# To add a thumbnail to the video element, use `thumbnail`
embed_video(mp4, thumbnail = png)

# To add placeholder text for non-HTML documents, use `placeholder`
embed_video(mp4, placeholder = "This is placeholder text.")

## Not run: 
# embed_video() is intended to be used in R Markdown code chunks or inline
# expressions. The following creates and knits an R Markdown document to
# HTML and PDF in your current working directory for you to inspect:
library(rmarkdown)
writeLines(c("# Hello embedr!",
"```{r embed-video, echo=TRUE}",
"embed_video(mp4, thumbnail = png, placeholder = 'This is placeholder text.')",
"```"), "test.Rmd")
render("test.Rmd", output_format = c('html_document', 'pdf_document'))

# Delete test files created by example code
unlink(c("test.Rmd", "test.html", "test.pdf"))

## End(Not run)

Enrich RTTM files

Description

Enrich Rich Transcription Time Marked (RTTM) files obtained from 'voice::read_rttm'.

Usage

enrich_rttm(listRttm, silence.gap = 0.5, as.tibble = TRUE)

Arguments

Value

A list containing either data frames or tibbles obtained from standard RTTM files. See 'voice::read_rttm'.

References

https://www.nist.gov/system/files/documents/itl/iad/mig/KWS15-evalplan-v05.pdf

See Also

voice::read_rttm

Examples

library(voice)

url0 <- 'https://raw.githubusercontent.com/filipezabala/voiceAudios/main/rttm/sherlock0.rttm'
destfile0 <- paste0(tempdir(), '/sherlock0.rttm')
download.file(url0, destfile = destfile0)
url1 <- 'https://raw.githubusercontent.com/filipezabala/voiceAudios/main/rttm/sherlock1.rttm'
destfile1 <- paste0(tempdir(), '/sherlock1.rttm')
download.file(url0, destfile = destfile1)

rttm <- voice::read_rttm(dirname(destfile0))
(er <- voice::enrich_rttm(rttm))
class(er)
lapply(er, class)

Expand model

Description

Expand model given y and x variables.

Usage

expand_model(y, x, k)

Arguments

Value

A char vector containing the expanded models.

Examples

library(voice)

expand_model('y', LETTERS[1:4], 1)
expand_model('y', LETTERS[1:4], 2)
expand_model('y', LETTERS[1:4], 3)
expand_model('y', LETTERS[1:4], 4)

# multiple models using apply functions
nx <- 10 # number of X variables to be used
models <- lapply(1:nx, expand_model, y = 'y', x = LETTERS[1:nx])
names(models) <- 1:nx
models
sum(sapply(models, length)) # total of models

Extract audio features

Description

Extracts features from WAV audio files.

Usage

extract_features(
  x,
  features = c("f0", "fmt", "gain"),
  filesRange = NULL,
  sex = "u",
  windowShift = 10,
  numFormants = 8,
  numcep = 12,
  dcttype = c("t2", "t1", "t3", "t4"),
  fbtype = c("mel", "htkmel", "fcmel", "bark"),
  resolution = 40,
  usecmp = FALSE,
  mc.cores = 1,
  full.names = TRUE,
  recursive = FALSE,
  check.mono = FALSE,
  stereo2mono = FALSE,
  overwrite = FALSE,
  freq = 44100,
  round.to = NULL,
  verbose = FALSE,
  pycall = "~/miniconda3/envs/pyvoice/bin/python"
)

Arguments

Details

The feature 'df' corresponds to 'formant dispersion' (df2:df8) by Fitch (1997), 'pf' to formant position' (pf1:pf8) by Puts, Apicella & Cárdena (2011), 'rf' to 'formant removal' (rf1:rf8) by Zabala (2023), 'rcf' to 'formant cumulated removal' (rcf2:rcf8) by Zabala (2023) and 'rpf' to 'formant position removal' (rpf2:rpf8) by Zabala (2023). The 'fmt_praat' feature may take long time processing. The following features may contain a variable number of columns: 'cep', 'dft', 'css' and 'lps'.

Value

A Media data frame containing the selected features.

References

Levinson N. (1946). The Wiener (root mean square) error criterion in filter design and prediction. Journal of Mathematics and Physics, 25(1-4), 261–278. (doi:10.1002/SAPM1946251261)

Durbin J. (1960). “The fitting of time-series models.” Revue de l’Institut International de Statistique, pp. 233–244. (https://www.jstor.org/stable/1401322)

Cooley J.W., Tukey J.W. (1965). “An algorithm for the machine calculation of complex Fourier series.” Mathematics of computation, 19(90), 297–301. (https://www.ams.org/journals/mcom/1965-19-090/S0025-5718-1965-0178586-1/)

Wasson D., Donaldson R. (1975). “Speech amplitude and zero crossings for automated identification of human speakers.” IEEE Transactions on Acoustics, Speech, and Signal Processing, 23(4), 390–392. (https://ieeexplore.ieee.org/document/1162690)

Allen J. (1977). “Short term spectral analysis, synthesis, and modification by discrete Fourier transform.” IEEE Transactions on Acoustics, Speech, and Signal Processing, 25(3), 235– 238. (https://ieeexplore.ieee.org/document/1162950)

Schäfer-Vincent K. (1982). "Significant points: Pitch period detection as a problem of segmentation." Phonetica, 39(4-5), 241–253. (doi:10.1159/000261665 )

Schäfer-Vincent K. (1983). "Pitch period detection and chaining: Method and evaluation." Phonetica, 40(3), 177–202. (doi:10.1159/000261691)

Ephraim Y., Malah D. (1984). “Speech enhancement using a minimum-mean square error short-time spectral amplitude estimator.” IEEE Transactions on acoustics, speech, and signal processing, 32(6), 1109–1121. (https://ieeexplore.ieee.org/document/1164453)

Delsarte P., Genin Y. (1986). “The split Levinson algorithm.” IEEE transactions on acoustics, speech, and signal processing, 34(3), 470–478. (https://ieeexplore.ieee.org/document/1164830)

Jackson J.C. (1995). "The Harmonic Sieve: A Novel Application of Fourier Analysis to Machine Learning Theory and Practice." Technical report, Carnegie-Mellon University Pittsburgh PA Schoo; of Computer Science.

Fitch, W.T. (1997) "Vocal tract length and formant frequency dispersion correlate with body size in rhesus macaques." J. Acoust. Soc. Am. 102, 1213 – 1222. (doi:10.1121/1.421048)

Boersma P., van Heuven V. (2001). Praat, a system for doing phonetics by computer. Glot. Int., 5(9/10), 341–347. (https://www.fon.hum.uva.nl/paul/papers/speakUnspeakPraat_glot2001.pdf)

Ellis DPW (2005). “PLP and RASTA (and MFCC, and inversion) in Matlab.” Online web resource.

Puts, D.A., Apicella, C.L., Cardenas, R.A. (2012) "Masculine voices signal men's threat potential in forager and industrial societies." Proc. R. Soc. B Biol. Sci. 279, 601–609. (doi:10.1098/rspb.2011.0829)

Examples

library(voice)

# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern = glob2rx('*.wav'), full.names = TRUE)

# minimal usage
M1 <- extract_features(path2wav)
M2 <- extract_features(dirname(path2wav))
identical(M1,M2)
table(basename(M1$wav_path))

# limiting filesRange
M3 <- extract_features(path2wav, filesRange = 3:6)
table(basename(M3$wav_path))

Features summary

Description

Returns summary measures of 'voice::extract_features'.

Usage

feat_summary(
  x,
  groupBy = "wav_path",
  wavPath = unique(x$wav_path),
  wavPathName = "wav_path",
  features = "f0",
  filesRange = NULL,
  sex = "u",
  windowShift = 10,
  numFormants = 8,
  numcep = 12,
  dcttype = c("t2", "t1", "t3", "t4"),
  fbtype = c("mel", "htkmel", "fcmel", "bark"),
  resolution = 40,
  usecmp = FALSE,
  mc.cores = 1,
  full.names = TRUE,
  recursive = FALSE,
  check.mono = FALSE,
  stereo2mono = FALSE,
  overwrite = FALSE,
  freq = 44100,
  round.to = 4,
  verbose = FALSE
)

Arguments

Details

filesRange should only be used when all the WAV files are in the same folder.

Value

A tibble data frame containing summarized numeric columns using (1) mean, (2) standard deviation, (3) variation coefficient, (4) median, (5) interquartile range and (6) median absolute deviation.

Examples

library(voice)

# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern = glob2rx('*.wav'), full.names = TRUE)

# creating Extended synthetic data
E <- dplyr::tibble(subject_id = c(1,1,1,2,2,2,3,3,3),
wav_path = path2wav)

# minimal usage
feat_summary(E)

# canonical data
feat_summary(E, groupBy = 'subject_id')

Get bit rate

Description

Get bit rate from WAV file.

Usage

get_bit(x)

Arguments

Value

Integer indicating the bit rate from a WAV file.

Examples

library(voice)

# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern <- glob2rx('*.wav'), full.names = TRUE)

rw <- tuneR::readWave(path2wav[1])
voice::get_bit(rw)

rwl <- lapply(path2wav, tuneR::readWave)
sapply(rwl, voice::get_bit)

Time duration

Description

Get time duration from WAV file.

Usage

get_dur(x)

Arguments

Value

Numeric indicating the time duration in seconds from a WAV file.

Examples

library(voice)

# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern <- glob2rx('*.wav'), full.names = TRUE)

rw <- tuneR::readWave(path2wav[1])
voice::get_dur(rw)

rwl <- lapply(path2wav, tuneR::readWave)
sapply(rwl, voice::get_dur)

Get left channel

Description

Get left channel from WAV file.

Usage

get_left(x)

Arguments

Value

Numeric vector indicating the left channel from a WAV file.

Examples

library(voice)

# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern <- glob2rx('*.wav'), full.names = TRUE)

rw <- tuneR::readWave(path2wav[1])
l <- voice::get_left(rw)
head(l)
length(l)

Get right channel

Description

Get right channel from WAV file.

Usage

get_right(x)

Arguments

Value

Numeric vector indicating the right channel from a WAV file.

Examples

library(voice)

# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern <- glob2rx('*.wav'), full.names = TRUE)

rw <- tuneR::readWave(path2wav[1])
r <- voice::get_right(rw)
head(r)
length(r)

Get sample rate

Description

Get sample rate from WAV file.

Usage

get_samp.rate(x)

Arguments

Value

Integer indicating the sample rate from a WAV file.

Examples

library(voice)

# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern <- glob2rx('*.wav'), full.names = TRUE)

rw <- tuneR::readWave(path2wav[1])
voice::get_samp.rate(rw)

rwl <- lapply(path2wav, tuneR::readWave)
sapply(rwl, voice::get_samp.rate)

Time beginning

Description

Get time beginning from a data frame in RTTM standard.

Usage

get_tbeg(x)

Arguments

Value

Numeric vector containing the time beginning in seconds.

Examples

library(voice)

url0 <- 'https://raw.githubusercontent.com/filipezabala/voiceAudios/main/rttm/sherlock0.rttm'
download.file(url0, destfile = paste0(tempdir(), '/sherlock0.rttm'))

rttm <- voice::read_rttm(tempdir())
(gtb <- voice::get_tbeg(rttm$sherlock0.rttm))
class(gtb)

Time duration

Description

Get time duration from a data frame in RTTM standard.

Usage

get_tdur(x)

Arguments

Value

Numeric vector containing the time duration in seconds.

Examples

library(voice)

url0 <- 'https://raw.githubusercontent.com/filipezabala/voiceAudios/main/rttm/sherlock0.rttm'
download.file(url0, destfile = paste0(tempdir(), '/sherlock0.rttm'))

rttm <- voice::read_rttm(tempdir())
(gtd <- voice::get_tdur(rttm$sherlock0.rttm))
class(gtd)

Interpolate vectors

Description

Interpolate vactors, compressing to compact.to fraction. May remove zeros.

Usage

interp(
  y,
  compact.to,
  drop.zeros = FALSE,
  to.data.frame = FALSE,
  round.off = NULL,
  weight = NULL
)

Arguments

Value

A list of interpolated x and y values with length near to compact.to*length(y).

See Also

rm0, interp_mc, interp_df

Examples

library(voice)

v1 <- 1:100
(c1 <- interp(v1, compact.to = 0.2))
length(c1$y)
plot(1:100, type = 'l')
points(c1$x, c1$y, col='red')

# with weight
(c2 <- interp(v1, compact.to = 0.2, weight = rev(v1)))
plot(c1$y)
points(c2$y, col = 'red')

(v2 <- c(1:5, rep(0,10), 1:10, rep(0,5), 10:20, rep(0,10)))
length(v2)
interp(v2, 0.1, drop.zeros = TRUE, to.data.frame = FALSE)
interp(v2, 0.1, drop.zeros = TRUE, to.data.frame = TRUE)
interp(v2, 0.2, drop.zeros = TRUE)
interp(v2, 0.2, drop.zeros = FALSE)

(v3 <- c(rep(0,10), 1:20, rep(0,3)))
(c3 <- interp(v3, 1/3, drop.zeros = FALSE, to.data.frame = FALSE))
lapply(c3, length)
plot(v3, type = 'l')
points(c3$x, c3$y, col = 'red')

(v4 <- c(rnorm(1:100)))
(c4 <- interp(v4, 1/4, round.off = 3))

Inperpolate data frames

Description

Interpolate data frames using multicore, compressing to compact.to fraction. May remove zeros.

Usage

interp_df(
  x,
  compact.to,
  id = colnames(x)[1],
  colnum = NULL,
  drop.x = TRUE,
  drop.zeros = FALSE,
  to.data.frame = TRUE,
  round.off = NULL,
  weight = NULL,
  mc.cores = 1
)

Arguments

Value

A data frame of interpolated values with nrow near to compact.to*length(x).

See Also

interp, interp_mc

Examples

library(voice)

# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern = glob2rx('*.wav'), full.names = TRUE)

# getting Media data frame via lean call
M <- extract_features(dirname(path2wav), features = c('f0','fmt'),
mc.cores = 1, verbose = FALSE)


(cM.df <- interp_df(M[,-(1:2)], 0.1, mc.cores = 1))
(cM.df2 <- interp_df(M[,-(1:2)], 0.1, drop.x = FALSE, mc.cores = 1))

dim(M)
dim(cM.df)
dim(cM.df2)
(cM.list <- interp_df(M[,-(1:2)], 0.1, to.data.frame = FALSE, mc.cores = 1))

Interpolate vectors using multicore

Description

Interpolate vectors using multicore

Usage

interp_mc(
  y,
  compact.to,
  drop.zeros = FALSE,
  to.data.frame = FALSE,
  round.off = NULL,
  weight = NULL,
  mc.cores = 1
)

Arguments

Value

A list of x and y convoluted values with length near to compact.to*length(y).

See Also

rm0, interp, interp_df

Examples

library(voice)
# Same result of interp() function if x is a vector
interp(1:100, compact.to = 0.1, drop.zeros = TRUE, to.data.frame = FALSE)
interp_mc(1:100, compact.to = 0.1, drop.zeros = TRUE, to.data.frame = FALSE)

interp(1:100, compact.to = 0.1, drop.zeros = TRUE, to.data.frame = TRUE)
interp_mc(1:100, compact.to = 0.1, drop.zeros = TRUE, to.data.frame = TRUE)

# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern = glob2rx('*.wav'), full.names = TRUE)


# getting Media data frame
M <- voice::extract_features(dirname(path2wav), mc.cores = 1, verbose = FALSE)

M.num <- M[,-(1:3)]
nrow(M.num)
cm1 <- interp_mc(M.num, compact.to = 0.1, drop.zeros = TRUE,
to.data.frame = FALSE, mc.cores = 1)
names(cm1)
lapply(cm1$f0, length)

Match string for audio suffix

Description

Given a character vector, returns a logical vector indicating which elements have a valid audio file extension.

Usage

is.audio(x)

Arguments

Return strings with a URL scheme

Description

Given a character vector, returns a logical indicating whether the URLs in the vector respond without error.

Usage

is.hosted(x)

Arguments

Return strings without a URL scheme

Description

Given a character vector, returns a logical indicating whether the paths in the vector point to existing local files.

Usage

is.local(x)

Arguments

Match string for URL prefix

Description

Given a character vector, returns a logical vector indicating which elements have a URL scheme.

Usage

is.url(x)

Arguments

Match string for video suffix

Description

Given a character vector, returns a logical vector indicating which elements have a valid video file extension.

Usage

is.video(x)

Arguments

Verify if an audio is mono

Description

Verify if an audio is mono

Usage

is_mono(x)

Arguments

Value

Logical. 'TRUE' indicates a mono (one-channel) file. 'FALSE' indicates a non-mono (two-channel) file.

Examples

library(voice)

# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern = glob2rx('*.wav'), full.names = TRUE)

is_mono(path2wav[1])
sapply(path2wav, is_mono)

These functions are sourced from the 'embedr' package by Michael McCarthy, under MIT License: https://github.com/mccarthy-m-g/embedr/blob/master/LICENSE.md This inclusion is temporary and will be discontinued once 'embedr' is available on CRAN. See https://github.com/mccarthy-m-g/embedr for more details.

Description

Example Media Files

Usage

mp3

Format

An object of class character of length 1.

Details

Example media files included with embedr.

- 'mp3': MP3 audio - 'mp4': MP4 video - 'png': PNG thumbnail

Sample IDs and paths

Description

A dataset containing sample IDs and paths from Ardila et al (2019) 'Common voice: A massively-multilingual speech corpus', used in Zabala (2023) 'voice: new approaches to audio analysis'. The considered sample contains 34,425 rows associated with 838 IDs (p_s = 2.4%).

Usage

mozilla_id_path

References

Ardila R, Branson M, Davis K, Henretty M, Kohler M, Meyer J, Morais R, Saunders L, Tyers FM, Weber G (2019). "Common voice: A massively-multilingual speech corpus." arXiv preprint arXiv:1912.06670. URL https://arxiv.org/abs/1912.06670.

See Also

extract_features.

Examples

library(voice)
mozilla_id_path

Assign notes to frequencies

Description

Returns a vector of notes for equal-tempered scale, A4 = 440 Hz.

Usage

notes(x, method = "spn", moving.average = FALSE, k = 11)

Arguments

Details

The symbol '#' is being used to represent a sharp note, the higher in pitch by one semitone on Scientific Pitch Notation (SPN).

Value

A vector containing the notes for equal-tempered scale, A4 = 440 Hz. When ‘method = ’spn'‘ the vector is of class ’ordered factor'. When ‘method = ’octave'‘ the vector is of class ’factor'. When ‘method = ’midi'‘ the vector is of class ’integer'.

References

https://gist.github.com/nvictor/7b4ab7070e210bc1306356f037226dd9

See Also

notes_freq

Examples

library(voice)
notes(c(220,440,880))
notes(c(220,440,880), method = 'octave')
notes(c(220,440,880), method = 'midi')

Frequencies on Scientific Pitch Notation (SPN)

Description

Returns a tibble of frequencies on Scientific Pitch Notation (SPN) for equal-tempered scale, A4 = 440 Hz.

Usage

notes_freq()

Details

The symbol '#' is being used to represent a sharp note, the higher in pitch by one semitone. The SPN is also known as American Standard Pitch Notation (ASPN) or International Pitch Notation (IPN).

Value

A tibble with frequencies for equal-tempered scale, A4 = 440 Hz.

References

https://gist.github.com/nvictor/7b4ab7070e210bc1306356f037226dd9

See Also

notes

Examples

library(voice)
notes_freq()

Piano plot

Description

Piano plot showing the notes in Scientific Pitch Notation.

Usage

piano_plot(data, num_fmt = 0)

Arguments

References

https://en.wikipedia.org/wiki/12_equal_temperament

https://en.wikipedia.org/wiki/Scientific_pitch_notation

Examples

library(voice)
# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern = glob2rx('*.wav'), full.names = TRUE)
# Media dataset
M <- extract_features(path2wav)
piano_plot(M, 0)
piano_plot(M, 0:2)

Read RTTM files

Description

Read Rich Transcription Time Marked (RTTM) files in fromRttm directory.

Usage

read_rttm(fromRttm)

Arguments

Details

The Rich Transcription Time Marked (RTTM) files are space-delimited text files containing one turn per line defined by NIST - National Institute of Standards and Technology. Each line containing ten fields:

type Type: segment type; should always by SPEAKER.

file File ID: file name; basename of the recording minus extension (e.g., rec1_a).

chnl Channel ID: channel (1-indexed) that turn is on; should always be 1.

tbeg Turn Onset – onset of turn in seconds from beginning of recording.

tdur Turn Duration – duration of turn in seconds.

ortho Orthography Field – should always by <NA>.

stype Speaker Type – should always be <NA>.

name Speaker Name – name of speaker of turn; should be unique within scope of each file.

conf Confidence Score – system confidence (probability) that information is correct; should always be <NA>.

slat Signal Lookahead Time – should always be <NA>.

Value

A list containing data frames obtained from standard RTTM files. See 'Details'.

References

https://www.nist.gov/system/files/documents/itl/iad/mig/KWS15-evalplan-v05.pdf

See Also

voice::enrich_rttm

Examples

library(voice)

url0 <- 'https://raw.githubusercontent.com/filipezabala/voiceAudios/main/rttm/sherlock0.rttm'
download.file(url0, destfile = paste0(tempdir(), '/sherlock0.rttm'))
url1 <- 'https://raw.githubusercontent.com/filipezabala/voiceAudios/main/rttm/sherlock1.rttm'
download.file(url0, destfile = paste0(tempdir(), '/sherlock1.rttm'))

(rttm <- voice::read_rttm(tempdir()))
class(rttm)
lapply(rttm, class)

Compress zeros.

Description

Transforms n sets of m>n zeros (alternated with sets of non zeros) into n sets of n zeros.

Usage

rm0(y)

Arguments

Value

Vector with n zeros.

Examples

library(voice)

(v0 <- c(1:20,rep(0,10)))
(r0 <- rm0(v0))
length(v0)
length(r0)
sum(v0 == 0)

(v1 <- c(rep(0,10),1:20))
(r1 <- rm0(v1))
length(r1)

(v2 <- rep(0,10))
(r2 <- rm0(v2))
length(r2)

(v3 <- c(0:10))
(r3 <- rm0(v3))
length(r3)

(v4 <- c(rep(0,10), 1:10, rep(0,5), 10:20, rep(0,10)))
(r4 <- rm0(v4))
length(r4)
sum(v4 == 0)

Smooth numeric variables in a data frame

Description

Smooth numeric variables in a data frame

Usage

smooth_df(x, k = 11, id = colnames(x)[1], colnum = NULL, mc.cores = 1)

Arguments

Value

Vector of interpolated values with length near to compact.to*length(x).

See Also

extract_features

Examples

library(voice)

# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern = glob2rx('*.wav'), full.names = TRUE)

# minimal usage
M <- extract_features(path2wav, features = c('f0', 'fmt'))
(Ms <- smooth_df(M[-(1:2)]))
dim(M)
dim(Ms)

Split Wave

Description

Split WAV files either in fromWav directory or using (same names) RTTM files/subdirectories as guidance.

Usage

splitw(
  fromWav,
  fromRttm = NULL,
  toSplit = NULL,
  autoDir = FALSE,
  subDir = FALSE,
  output = "wave",
  filesRange = NULL,
  full.names = TRUE,
  recursive = FALSE,
  silence.gap = 0.5
)

Arguments

Details

When autoDir = TRUE, the following directories are created: '../mp3','../rttm', '../split' and '../musicxml'. Use getwd() to find the parent directory '../'.

Value

Splited audio files according to the correspondent RTTM file(s). See 'voice::diarize'.

See Also

voice::diarize

Examples

## Not run: 
library(voice)

urlWav <- 'https://raw.githubusercontent.com/filipezabala/voiceAudios/main/wav/sherlock0.wav'
destWav <- paste0(tempdir(), '/sherlock0.wav')
download.file(urlWav, destfile = destWav)

urlRttm <- 'https://raw.githubusercontent.com/filipezabala/voiceAudios/main/rttm/sherlock0.rttm'
destRttm <- paste0(tempdir(), '/sherlock0.rttm')
download.file(urlRttm, destfile = destRttm)

splitDir <- paste0(tempdir(), '/split')
dir.create(splitDir)
splitw(destWav, fromRttm = destRttm, toSplit = splitDir)

dir(splitDir)

## End(Not run)

Convert SPN to ABC

Description

Convert SPN to standard octave.

Usage

spn2abc(x, to_lower = FALSE, spacing = TRUE)

Arguments

References

https://en.wikipedia.org/wiki/Scientific_pitch_notation

https://en.wikipedia.org/wiki/ABC_notation

Examples

library(voice)
spn2abc('C4')
spn2abc('C5')
spn2abc('C4', to_lower = TRUE)
spn2abc(c('C4','D#7','E2'))
spn2abc(c('C4','D#7','E2'), to_lower = TRUE)
spn2abc(c('C4','D#7','E2'), spacing = FALSE)
spn2abc(c('C4','D#7','E2'), to_lower = TRUE, spacing = FALSE)

Tag a data frame with media information

Description

Tag a data frame with media information

Usage

tag(
  x,
  groupBy = "wav_path",
  wavPath = unique(x$wav_path),
  wavPathName = "wav_path",
  tags = c("feat_summary"),
  sortByGroupBy = TRUE,
  filesRange = NULL,
  features = "f0",
  sex = "u",
  windowShift = 5,
  numFormants = 8,
  numcep = 12,
  dcttype = c("t2", "t1", "t3", "t4"),
  fbtype = c("mel", "htkmel", "fcmel", "bark"),
  resolution = 40,
  usecmp = FALSE,
  mc.cores = 1,
  full.names = TRUE,
  recursive = FALSE,
  check.mono = FALSE,
  stereo2mono = FALSE,
  overwrite = FALSE,
  freq = 44100,
  round.to = 4,
  verbose = FALSE
)

Arguments

Details

filesRange should only be used when all the WAV files are in the same folder.

Value

A tibble data frame containing summarized numeric columns using (1) mean, (2) standard deviation, (3) variation coefficient, (4) median, (5) interquartile range and (6) median absolute deviation.

Examples

library(voice)

# get path to audio file
path2wav <- list.files(system.file('extdata', package = 'wrassp'),
pattern = glob2rx('*.wav'), full.names = TRUE)

# creating Extended synthetic data
E <- dplyr::tibble(subject_id = c(1,1,1,2,2,2,3,3,3),
wav_path = path2wav)
E

# minimal usage
tag(E)

# canonical data
tag(E, groupBy = 'subject_id')

# limiting filesRange
tag(E, filesRange = 3:6)

# more features
Et <- tag(E, features = c('f0', 'fmt', 'rf', 'rcf', 'rpf', 'rfc', 'mfcc'),
groupBy = 'subject_id')
Et
str(Et)

These functions are sourced from the 'embedr' package by Michael McCarthy, under MIT License: https://github.com/mccarthy-m-g/embedr/blob/master/LICENSE.md This inclusion is temporary and will be discontinued once 'embedr' is available on CRAN. See https://github.com/mccarthy-m-g/embedr for more details.

Description

Check if URL exists

Usage

url.exists(x)

Arguments

Details

Given a character string, returns a logical vector indicating whether a request for a specific URL responds without error.

Value

'TRUE' if the URL responds without error, otherwise 'FALSE'.

Writes a list to a path

Description

Writes a list to a path

Usage

write_list(x, path)

Arguments

Value

A file named 'list.txt' in 'path'.

Examples

## Not run: 
library(voice)

pts <- list(x = cars[,1], y = cars[,2])
listFile <- paste0(tempdir(), '/list.txt')
voice::write_list(pts, listFile)
file.info(listFile)
system(paste0('head ', listFile))

## End(Not run)