Skip to contents

Track fundamental frequency using the KSV periodicity detector

trk_pitch_ksv.Rd

Estimates the fundamental frequency fo using the Schäfer-Vincent periodicity detection algorithm (Schäfer-Vincent 1983) implemented in the libassp C library (Scheffers 2012) . This extremum-based method is fast and works directly on the waveform without spectral analysis.

Usage

trk_pitch_ksv(listOfFiles, beginTime = 0, endTime = 0, windowShift = 5, gender = "u", maxF = 600, minF = 50, minAmp = 50, maxZCR = 3000, toFile = FALSE, explicitExt = "fo", outputDirectory = NULL, assertLossless = NULL, logToFile = FALSE, convertOverwrites = FALSE, keepConverted = FALSE, verbose = TRUE)

Arguments

listOfFiles

Character vector of audio file paths. Any format supported by av is accepted; non-native inputs are transcoded automatically.

beginTime

Numeric. Start of analysis window in seconds. Default 0 (file start).

endTime

Numeric. End of analysis window in seconds. Default 0 (file end).

windowShift

Numeric. Frame shift in milliseconds; sets output frame rate (1000 / windowShift Hz). Default 5 ms.

gender

Character. Gender-specific fo search range: "f" (female, 80–640 Hz), "m" (male, 50–400 Hz), "u" (unknown, 50–600 Hz, default).

maxF

Numeric. Maximum fo in Hz. Default 600.

minF

Numeric. Minimum fo in Hz. Default 50.

minAmp

Numeric. Minimum waveform amplitude threshold for voiced frames. Default 50.

maxZCR

Numeric. Maximum zero-crossing rate in Hz for voicing detection. Default 3000.0.

toFile

Logical. If TRUE, write SSFF output files and return the count written (invisibly). If FALSE, return an AsspDataObj. Default FALSE.

explicitExt

Character. Output file extension. Default "fo".

outputDirectory

Character. Directory for output files. NULL (default) writes alongside the input file.

assertLossless

Character vector of additional file extensions to treat as losslessly encoded.

logToFile

Logical. Write processing log to a file in outputDirectory rather than the console. Default FALSE.

convertOverwrites

Logical. Allow transcoding to overwrite existing files. Default FALSE.

keepConverted

Logical. Retain intermediate transcoded files. Default FALSE.

verbose

Logical. Print per-file progress. Default FALSE.

Value

If toFile = FALSE: an AsspDataObj with track:

fo[Hz]

REAL32, Hz, n_frames x 1 column. Estimated fundamental frequency; 0 indicates unvoiced frames.

Frame rate: 1000 / windowShift Hz (default 200 Hz). If toFile = TRUE: integer count of files written, returned invisibly.

Details

gender sets the default minF/maxF search range but is overridden by explicit minF/maxF values. minAmp and maxZCR control voicing detection independently of the pitch range.

References

Schäfer-Vincent K (1983). “Pitch Period Detection and Chaining: Method and Evaluation.” Phonetica, 40(3), 177–202. ISSN 0031-8388. doi:10.1159/000261691 .

Scheffers M (2012). “Advanced Speech Signal Processor.” https://sourceforge.net/projects/libassp/files/libassp/.

See also

wrassp::ksvF0

Author

Raphael Winkelmann

Lasse Bombien

Fredrik Nylén

Examples

# get path to audio file
path2wav <- list.files(system.file("samples","sustained", package = "superassp"), pattern = glob2rx("a1.wav"), full.names = TRUE)

# calculate fundamental frequency contour
res <- trk_pitch_ksv(path2wav, toFile=FALSE)
#> Applying `method(trk_pitch_ksv, class_character)()` to 1 recording

# plot the fundamental frequency contour
plot(seq(0, n_records(res) - 1) / sample_rate(res) +
      attr(res, 'startTime'),
    res[["fo[Hz]"]],
    type='l',
    xlab='time (s)',
    ylab=expression(paste(f[o]," frequency (Hz)")))