Main Page | Download | Code | Examples | Documentation | MFAQ | Roadmap

Command line parameters

versions concerned : 0.2.3

--help, -h, /?

Display help, i.e. a list of command line arguments with a short explanation.


Display a list of more "advanced" command line arguments with a short explanation.

--version, -v

Display the version of the program as well as the last code modification date.

--quiet, -q

No-prompt mode, meaning the program won't ever ask you for any direct input. Useful for scripting.

--analysis, -a

Analysis mode. Implicit if the input file's name ends in '.wav'.

--sine, -s

Sine synthesis mode.

--noise, -n

Noise synthesis mode.

Note : Read paragraph 'II. Mode selection' of the Basic Operation document in order to decide which of the two synthesis modes above to use.
Also, please refer to paragraph 'III. Processing parameters' of the same document for directions on how to use the following parameters.

--min-freq, -min [real]

Minimum frequency in Hertz. Can only be 0 or even negative if the logarithmic base is less than 2, i.e. for example in linear mode.

--max-freq, -max [real]

Maximum frequency in Hertz. Note that it's OK if it goes beyond the Nyquist frequency (the highest frequency that can be present in a digitised sound) regardless of the operation mode.

--bpo, -b [real]

Frequency resolution in Bands Per Octave. This setting should be avoided for logarithmic scales other than 2.

--pps, -p [real]

Time resolution in Pixels Per Second.

--height, -y [integer]

Specifies the desired height of the spectrogram

--width, -x [integer]

Specifies the desired width of the spectrogram. The actual output size may be 1 pixel shorter or longer.

--sample-rate, -r [integer]

Sample rate of the output sound. Must be specified in synthesis mode, the recommended value for most sounds being 44100.

--brightness, -g [real]

Brightness of the image, a bit like gamma correction. Applies the formula f(x) = x^1/brightness to the output image, and f(x) = x^brightness to the input image. The default value is 1, giving the image a linear intensity scale and a dynamic range of 48 dB, but because this is usually very dark I often use a value of 2, which is equivalent to a square root on the output/square on the input, and extend the dynamic range to about 96 dB. The dynamic range of the spectrogram is given by brightness * 48 dB.

--format-param, -f [integer]

Output format option. This is bit-depth for WAV files, and there is no option for BMP files as there's only one supported BMP format. Three options are supported for WAV files :

--log-base [real]

Frequency scale's logarithmic base. Default is 2, which is arguably best suited for most sounds, but most particularly music. A value of 1 is handled as an exception as a linear scale. It's the same thing as --linear or -l.

--linear, -l

Linear frequency scale. Produces an output similar to most other available spectrographs. Same as --log-base 1.        (appeared in version 0.2.3)

--loop-size [real]

Noise look-up table size in seconds. Noise synthesis is based on modulation by filtered noises, based on a fixed-duration looping single noise. This parameter defines how long this loop lasts. As of version 0.2.3 a lower value will not make things much faster as the bottleneck in noise synthesis is envelope interpolation. Default is 10 (seconds).

--bmsq-lut-size [integer]

Blackman Square kernel LUT size. Some obscure setting related to envelope interpolation in noise synthesis, interpolation which is based on a kernel contained in a look-up table (LUT) which size is defined by this parameter. Default is 16000.

--pi [real]

Pi. Just so you can see how well things work if pi was 22/7 or even 3.2. Default is 3.1415926535897932 (enough digits for double precision floats).

This site is in hiatus. Last updated on February 23rd, 2009
©2007-2009 Michel Rouzic