Track LP-smoothed spectrum

Computes a short-term spectral envelope smoothed by linear prediction using the libassp C library (Scheffers 2012) . LP smoothing produces a parametric model-based envelope; prefer trk_lps_spectrum over trk_css_spectrum when the LP order is known and explicit control over the number of resonances is desired.

Usage

trk_lps_spectrum(
  listOfFiles,
  beginTime = 0,
  centerTime = FALSE,
  endTime = 0,
  resolution = 40,
  fftLength = 0,
  windowSize = 20,
  windowShift = 5,
  window = "BLACKMAN",
  order = 0,
  preemphasis = -0.95,
  deemphasize = TRUE,
  toFile = FALSE,
  explicitExt = "lps",
  outputDirectory = NULL,
  assertLossless = NULL,
  logToFile = FALSE,
  keepConverted = FALSE,
  convertOverwrites = 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).
centerTime: Numeric or logical. Single-frame analysis time point in seconds; overrides beginTime, endTime, and windowShift. Default FALSE.
endTime: Numeric. End of analysis window in seconds. Default 0 (file end).
resolution: Numeric. Target FFT frequency resolution in Hz. Default 40.0.
fftLength: Integer. Explicit FFT length in points; overrides resolution. Default 0 (use resolution).
windowSize: Numeric. Analysis window size in milliseconds. Default 20 ms.
windowShift: Numeric. Frame shift in milliseconds; sets output frame rate (1000 / windowShift Hz). Default 5 ms.
window: Character. Analysis window function type. Default "BLACKMAN". See AsspWindowTypes.
order: Integer. LP prediction order; 0 defaults to sample rate in kHz + 3. Default 0.
preemphasis: Numeric. Pre-emphasis factor applied before LP analysis. Default -0.95.
deemphasize: Logical. Undo the spectral tilt introduced by pre-emphasis when computing the output spectrum. Default TRUE.
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 "lps".
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.
keepConverted: Logical. Retain intermediate transcoded files. Default FALSE.
convertOverwrites: Logical. Allow transcoding to overwrite existing files. Default FALSE.
verbose: Logical. Print per-file progress. Default TRUE.

Value

If toFile = FALSE: an AsspDataObj with track:

LPS[dB]: REAL32, dB, n_frames x (FFT_length/2 + 1) columns. LP-smoothed spectral amplitude from 0 Hz to the Nyquist rate.

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

Details

order = 0 sets the LP order to sample rate in kHz + 3. When deemphasize = TRUE (default), the output spectrum is corrected for the pre-emphasis spectral tilt applied before LP analysis.

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 linear prediction smoothed spectrum
res <- trk_lps_spectrum(path2wav, toFile=FALSE)
#> Applying `method(trk_lps_spectrum, class_character)()` to 1 recording

# coefficients run from 0 Hz to the Nyquist rate
resolution <- attr(res,"origFreq") / (2 * (ncol(res[[1]]) - 1))

# plot spectral values at midpoint of signal
plot(y=res[["LPS[dB]"]][400,],
    x=seq(0,ncol(res[[1]]) - 1) * resolution,
    type='l',
    xlab='Frequency (Hz)',
    ylab='Amplitude (dB)')