Command-line utilities
loristrck provides a set of command-line utilities to analyze and transform spectra directly from the command-line
These utilities are installed alongside the loristrck python package
loristrck_analyze
Partial tracking analysis. Takes a soundfile as input and generates a SDIF file of 1TRC or RBEP format
usage: loristrck_analyze [OPTIONS] sndfile
Options
-r RESOLUTION/--resolution RESOLUTION: Only one partial will be found within this distance. Usual values range from 30 Hz to 200 Hz. As a rule of thumb, when tracking a monophonic source,resolution ~= min(f0) * 0.9. So if the source is a male voice dropping down to 70 Hz, resolution=60 Hz. The resolution determines the size of the fft-w WINSIZE/--winsize WINSIZE: Hz. Is the main lobe width of the Kaiser analysis window in Hz (main-lobe, zero to zero). If not given, a default value is calculated. In general window size should be higher than the resolution (default =resolution*2)--hoptime HOPTIME: The time to shift the window after each analysis (default: 1/windowsize, which results in 1x overlap). For 4x overlap and a window size of 60 Hz, hoptime should be 0.00417. See also--hopoverlap--hopoverlap OVERLAP: If given, sets the hoptime in terms of overlapping windows. A hop overlap of 2 will result in ahop time = 1/(2*windowsize)--ampfloor AMPFLOOR: min. amplitude of any breakpoint, in dB (default = -120)--freqdrift FREQ: Hz. The maximum variation of frecuency between two breakpoints to be considered to belong to the same partial. A sensible value is between 1/2 to 3/4 of resolution:freqdrift=0.62*resolution--sidelobe SIDELOBE: dB (default: 90 dB). A positive dB value, indicates the shape of the Kaiser window--residuebw RESIDUE: Hz (default = 2000 Hz). Construct Partial bandwidth env. by associating residual energy with the selected spectral peaks that are used to construct Partials. The bandwidth is the width (in Hz) association regions used. Defaults to 2 kHz, corresponding to 1 kHz region center spacing.--croptime TIME: sec. Max. time correction beyond which a reassigned bp is considered unreliable, and not eligible. Default: the hop time.--outfile OUTFILE: The generated sdif file--sdiftype SDIFTYPE: One of 'rbep' or '1trk' (case is ignored)--minbps MINBPS: Min. number of breakpoints for a partial to not be discarded--minamp MINAMP: Min. amplitude of a partial (in average) to not be discarded--fadetime FADETIME: fade time used when a partial does not end with a 0 amp breakpoint
loristrck_pack
Analyzes a soundfile or loads a previously analyzed sdiffile, packing the partials into a number of simultaneous streams, in order to be resynthesized.
The result is a matrix of at most maxoscil*3 number of columns, the height of the
matrix being dependent on the length of the sound and the sampling interval used.
The matrix is saved as an uncompressed .wav file with the format:
Header: [dataOffset=5, dt, numcols, numrows, t0]
And then the data, a matrix of numcols x numrows, where:
numcols = numstreams * 3
(each stream consists of three columns: freq, amp and bandwidth)
numrows = each row consists of a sample at time t, where t = t0 + row*dt
This is mostly thought to be used in tandem with csound's beadsynt opcode, which performs band-enhanced-additive-synthesis in realtime
Note
If you need more control over the analysis step, first analyze the soundfile to obtain a sdif file, then use this utility to pack a matrix (see loristrck_analyze)
positional arguments
--------------------
infile: An infile can be a soundfile or a sdif file
optional arguments
------------------
-h, --help show this help message and exit
-r RESOLUTION, --resolution RESOLUTION
Analysis resolution, in Hz (only for analysis)
-w WINSIZE, --winsize WINSIZE
Window size (in Hz) (only valid if doing analysis)
-o OUTFILE, --outfile OUTFILE
Name of the resulting wav file
--minbps MINBPS Min. number of breakpoints for a partial to be packed
--minamp MINAMP
--mindur MINDUR
--maxoscil MAXOSCIL default=infinite
--maxtracks MAXTRACKS
Max. number of tracks (leave as 0 if table size is not a problem)
--dt DT A sampling interval, in seconds. As a rule of thumb,
dt should be somewhat smaller than ksmps/sr
loristrck_synth
Synthesize a .sdif or .mtx file
A .sdif file is the result of performing analysis, and can be any RBEP or 1TRC sdif file,
as produced, for example, via loristrck_analyze
A .mtx file is a packed spectrum, where partials are packed in non-simultaneous tracks
to produce facilitate playback. See loristrck_pack for more information
usage: loristrck_synth [-h] [--speed SPEED] [--transposition TRANSPOSITION] [--noise {uniform,gaussian}]
[--quality QUALITY] [-o OUT]
inputfile
positional arguments:
inputfile A .sdif or .mtx file (as generated via loristrck_pack
optional arguments:
-h, --help show this help message and exit
--speed SPEED Playback speed. Pitch is not modified
--transposition TRANSPOSITION
Transposition in semitones (independent from playback speed)
--noise {uniform,gaussian}
Noise type used for the residual part when synthesizing a .mtx file. The original
implementation uses gaussian noise
--quality QUALITY Oscillator quality when playing a .mtx file. 0: fast, 1: fast + freq. interpolation, 2:
linear interpolation, 3: linear interpolation + freq. interpolation
-o OUT, --out OUT Play / Save the samples. Use dac to play in realtime (.mtx files only), or a .wav of .aif
path to synthesize to that file
loristrck_chord
Extract a chord from a soundfile at a given time
usage: loristrck_chord [OPTIONS] sdiffile
TODO
positional arguments:
sdiffile The input file, a sdif file with format 1TRC or RBEP
optional arguments:
-h, --help show this help message and exit
-t TIME, --time TIME Time to extract the chord
-n MAXCOUNT, --maxcount MAXCOUNT
Max. number of partials
-a MINAMP, --minamp MINAMP
Min. amplitude of a breakpoint to be considered (in dB)
-d MINDUR, --mindur MINDUR
Min. dur of a partial to be considered
--minfreq MINFREQ Min. freq of a partial
--maxfreq MAXFREQ Max. freq of a partial
--sortby {frequency,amplitude}
-s SOUND, --sound SOUND
Play / Generate a soundfile with the extracted chord.
Use 'dac' to play directly
--sounddur SOUNDDUR Duration of the generated sound output
--fade FADE
--a4 A4 Reference frequency (default = 442 Hz)