| Title: | Miscellaneous Convenience Functions and Dataset |
|---|---|
| Description: | Miscellaneous convenience functions and wrapper functions to convert frequencies between Hz, semitones, mel and Bark, to create a matrix of dummy columns from a factor, to determine whether x lies in range [a,b], and to add a bracketed line to an existing plot. This package also contains an example data set of a stratified sample of 80 talkers of Dutch. |
| Authors: | Hugo Quene [aut, cre] |
| Maintainer: | Hugo Quene <[email protected]> |
| License: | GPL (>= 2) |
| Version: | 0.2-1 |
| Built: | 2025-02-22 06:10:17 UTC |
| Source: | https://github.com/cran/hqmisc |
Create a matrix of dummy vectors (consisting of 0 and 1 only) from factor x. This may be useful if you want to use dummy predictors in regression analysis.
as.dummies ( x )as.dummies ( x )
x |
input factor. |
Levels of the input factor correspond with columns in the output matrix. The integer value in cell [i,j] of the returned matrix indicates whether (value=1) or not (value=0) the i'th observation in the input factor x has the value of the j'th level of that factor.
as.dummies(x) is similar to model.matrix(~-1+x).
Matrix of integer values 0 or 1. The number of rows of the returned matrix equals the length of the input factor, and the number of columns equals the number of levels of the input factor.
levels, matrix, model.matrix..
xf <- factor( rep(letters[1:3],each=2) ) as.dummies(xf)xf <- factor( rep(letters[1:3],each=2) ) as.dummies(xf)
Add a rectangular bracket to the most recent plot. This bracket may be used to loosely connect points or bars in a plot, to attract the viewer's attention to these connected points or bars.
bracket( x0, y0, x1=x0, y1=y0, offset=1, length=offset/2, side=1, col="grey", ... )bracket( x0, y0, x1=x0, y1=y0, offset=1, length=offset/2, side=1, col="grey", ... )
x0, y0
|
coordinates of points from which to draw bracket. |
x1, y1
|
coordinates of points to which to draw bracket. |
offset |
offset or distance of bracket from specified coordinates; the direction of the offset depends on side. The offset must be specified in data units of the appropriate dimension of the plot. |
length |
the length of the whiskers of the bracket. |
side |
the side, relative to the specified coordinates, on which to draw the bracket; 1=below, 2=left, 3=above, 4=right. |
col |
the color in which to draw the bracket. |
... |
other graphical parameters to be passed on (see par),
such as |
If offset is positive (the default), then positive values of length (the default) result in inside whiskers, towards the specified coordinates, and negative values of length result in outside whiskers, away from the specified coordinates.
The value for side must be specified as an integer number; values other than 1:4 will be ignored.
None; a bracket is added to the existing plot.
plot( 1:9, 1:9 ) bracket( x0=4, x1=6, y0=3, side=1 ) bracket( x0=3, y0=4, y1=6, side=2, offset=1, length=1, lty=2 ) bracket( x0=4, x1=6, y0=7, side=3, col="red" ) # negative length: whiskers outside bracket( x0=7, y0=4, y1=6, side=4, offset=0, length=-1 )plot( 1:9, 1:9 ) bracket( x0=4, x1=6, y0=3, side=1 ) bracket( x0=3, y0=4, y1=6, side=2, offset=1, length=1, lty=2 ) bracket( x0=4, x1=6, y0=7, side=3, col="red" ) # negative length: whiskers outside bracket( x0=7, y0=4, y1=6, side=4, offset=0, length=-1 )
Convert between frequency in Hz and in psychoacoustic Bark units
f2bark( hz ) bark2f( bark )f2bark( hz ) bark2f( bark )
hz |
a frequency in Hertz (cycles per second). |
bark |
a frequency in Bark units. |
These functions convert between the linear Hertz and psychoacoustic Bark scales. The functions use Formula (6) of Traunmüller (1990).
Transformed value(s) in Bark or Hertz units.
Traunmüller, H. (1990). Analytical expressions for the tonotopic sensory scale. Journal of the Acoustical Society of America, 88(1), 97-100.
f2bark( c(1,3,5)*500 ) bark2f( 12 )f2bark( c(1,3,5)*500 ) bark2f( 12 )
Convert between frequency in Hz and in psychoacoustic mel units.
f2mel( hz ) mel2f( mel )f2mel( hz ) mel2f( mel )
hz |
a frequency in Hertz (cycles per second). |
mel |
a frequency in mel units. |
These functions convert between the linear Hertz and psychoacoustic Mel scales. The functions use formulas provided by Rietveld & Van Heuven (2009).
A frequency of 1000 Hz equals 1000 mel. For frequencies below 1000 Hz the mel scale is approximately linear; for frequencies above 1000 Hz the mel scale is approximately logarithmic.
Transformed value(s) in mel or hertz units.
Rietveld, A. C. M., & Van Heuven, V. J. (2009). Algemene Fonetiek (3rd ed.). Bussum: Coutinho.
f2mel( seq( 250, 1000, by= 250) ) f2mel( seq(1000, 3000, by=1000) ) mel2f( seq( 500, 2500, by= 500) )f2mel( seq( 250, 1000, by= 250) ) f2mel( seq(1000, 3000, by=1000) ) mel2f( seq( 500, 2500, by= 500) )
Convert between frequency in Hz and in semitones, relative to baseline frequency in Hz.
f2st( hz, base=50 ) st2f( st, base=50 )f2st( hz, base=50 ) st2f( st, base=50 )
hz |
a frequency in Hertz (cycles per second). |
st |
a frequency in semitones relative to the base frequency. |
base |
the baseline frequency (in Hertz units) relative to which semitones are expressed. |
These functions convert between the linear Hertz and logarithmic semitone scales. A doubling of the frequency in Hz equals 12 semitones or 1 octave.
Function f2st was based on function h2st by Mark Liberman
(last seen at http://ldc.upenn.edu/myl/llog/semitones.R but no longer available online).
Function st2f was based on a conversion script by Jan Roelof de Pijper
(last seen at http://users.abo.fi/jtuomain/speech/semitone.html but no longer available online).
For frequencies in Hz of musical notes, see e.g. http://www.phy.mtu.edu/~suits/notefreqs.html.
Transformed value(s) in semitones or in Hertz (cycles per second).
# distance in semitones between target and baseline frequencies in Hz. f2st( 440, base=110 ) # frequency in Hz of a tone that is +24 semitones distant # from the baseline, i.e. a distance of +two octaves. st2f( 24 )# distance in semitones between target and baseline frequencies in Hz. f2st( 440, base=110 ) # frequency in Hz of a tone that is +24 semitones distant # from the baseline, i.e. a distance of +two octaves. st2f( 24 )
Check whether argument x is within range.
is.inrange( x, range=c(0,1) )is.inrange( x, range=c(0,1) )
x |
numerical value(s) to be evaluated. |
range |
list of two numbers defining the range used in the evaluation of x. |
The return value indicates whether x is an element of [range[1], range[2]].
Note that the range used in the evaluation includes both endpoints of the specified range.
Logical.
is.inrange(1/2) is.inrange(pi, range=c(0,pi))is.inrange(1/2) is.inrange(pi, range=c(0,pi))
This data set gives metadata (id, sex, age, region of origin) and speech characteristics (average syllable duration, average phrase length) for a stratified sample of 80 talkers of Dutch from the Netherlands.
data(talkers)data(talkers)
A data frame with 80 observations on the following 6 variables.
ididentifier code (from data source, see Source)
sexsex (0=female, 1=male)
ageage (in years)
regionregion of origin (a factor with levels M=Mid, N=North, S=South, or W=West)
sylduraverage duration of syllables, or seconds per syllable (in seconds, excluding pause time, 1/(articulation rate) )
nsylaverage number of syllables per phrase, or average phrase length in syllables
Talkers grew up in their region of origin, and have lived and worked there as teachers of Dutch Language and Literature in secondary education. Talkers with ages between 41 and 45 were not included in this study. The sample is stratified by sex, region, and (age>41) (see Examples).
Speech data were collected from (and averaged over) a recorded interview lasting about 15 minutes. The talker and the interviewer only spoke Standard Dutch during the interview.
One talker (id 117) spoke remarkably slower than all others, yielding a very high syldur.
The West region is commonly regarded as the linguistic center of the Netherlands. Each of the four regions has a distinct variety of Dutch. The variety of the West region is closest to the Standard Dutch spoken in the Netherlands.
Speech recordings and metadata were collected in 1999.
http://tla.mpi.nl/resources/data-archive/, Corpus of Spoken Dutch
Oostdijk, N. (2000). The Spoken Dutch Corpus: Overview and first evaluation. In M. Gravilidou, G. Carayannis, S. Markantonatou, S. Piperidis & G. Stainhaouer (Eds.), Proceedings of the Second International Conference on Language Resources and Evaluation (Vol. 2, pp. 887-894).
Adank, P., van Hout, R., & van de Velde, H. (2007). An acoustic description of the vowels of northern and southern Standard Dutch II: Regional varieties. Journal of the Acoustical Society of America, 121(2), 1130-1141.
Quené, H. (2008). Multilevel modeling of between-speaker and within-speaker variation in spontaneous speech tempo. Journal of the Acoustical Society of America, 123(2), 1104-1113.
data(talkers) str(talkers) pairs( talkers[,2:6] ) with( talkers, table( sex, region, I(age>41) ) )data(talkers) str(talkers) pairs( talkers[,2:6] ) with( talkers, table( sex, region, I(age>41) ) )