5. Quick reference guide

This chapter is meant as a reference guide for all the SPro tools, summarizing the syntax, synopsis and options. This is actually a printed version of the online help message obtained with `--help'.

5.1 sfbank		sfbank reference
5.2 sfbcep		sfbcep reference
5.3 slpc		slpc reference
5.4 slpcep		slpcep reference
5.5 scopy		scopy reference

5.1 sfbank

Usage		Syntax
Synopsis		Brief description
Options

Usage

Synopsis

Options

-F, --format=str

Specify the input waveform file format. Available formats are `PCM16', `wave' or `sphere'. Default: `PCM16'.

-f, --sample-rate=f

Set input waveform sample rate to f Hz for `PCM16' waveform files. Default: 8 kHz.

-x, --channel=n

Set the channel to consider for feature extraction. Default: 1.

-B, --swap

Swap the input waveform samples.

-I, --input-bufsize=n

Set the input buffer size to n kbytes. Default: 10 Mb.

-O, --output-bufsize=n

Set the output buffer size to n kbytes. Default: 10 Mb.

-k, --pre-emphasis=f

Set the pre-emphasis coefficient to f. Default: 0.95.

-l --length=f

Set the analysis frame length to f ms. Default: 20.0 ms.

-d, --shift=f

Set the interval between two consecutive frames to f ms. Default: 10.0 ms.

-w, --window=str

Specify the waveform weighting window. Available windows are `Hamming', `Hanning', `Blackman' or `none'. Default: `Hamming'.

-n, --num-filters=n

Set the number of channels in the filter bank. Default: 24.

-a, --alpha=f

Set the bilinear frequency warping factor to f. Default: 0.

-m, --mel

Use MEL frequency warping. Overwrites `--alpha'.

-i, --freq-min=f

Set the lower frequency bound to f Hz. Default: 0 Hz.

-u, --freq-max=f

Set the upper frequency bound to f Hz. Default: Niquist.

-b, --fft-length=n

Set FFT length to n samples. Default: 512.

-D, --delta

Add first order derivatives to the feature vector.

-A, --acceleration

Add second order derivatives to the feature vector. Requires `--delta'.

-v, --verbose

Turn on verbose mode

-h, --help

Print a help message for the tool and exit.

-V, --version

Print version information and exit.

5.2 sfbcep

Usage		Syntax
Synopsis		Brief description
Options

Usage

sfbcep [options] ifile ofile

Synopsis

Filter-bank based cepstral analysis of the input signal.

Options

-F, --format=str

Specify the input waveform file format. Available formats are `PCM16', `wave' or `sphere'. Default: `PCM16'.

-f, --sample-rate=f

Set input waveform sample rate to f Hz for `PCM16' waveform files. Default: 8 kHz.

-x, --channel=n

Set the channel to consider for feature extraction. Default: 1.

-B, --swap

Swap the input waveform samples.

-I, --input-bufsize=n

Set the input buffer size to n kbytes. Default: 10 Mb.

-O, --output-bufsize=n

Set the output buffer size to n kbytes. Default: 10 Mb.

-k, --pre-emphasis=f

Set the pre-emphasis coefficient to f. Default: 0.95.

-l --length=f

Set the analysis frame length to f ms. Default: 20.0 ms.

-d, --shift=f

Set the interval between two consecutive frames to f ms. Default: 10.0 ms.

-w, --window=str

Specify the waveform weighting window. Available windows are `Hamming', `Hanning', `Blackman' or `none'. Default: `Hamming'.

-n, --num-filters=n

Set the number of channels in the filter bank. Default: 24.

-a, --alpha=f

Set the bilinear frequency warping factor to f. Default: 0.

-m, --mel

Use MEL frequency warping. Overwrites `--alpha'.

-i, --freq-min=f

Set the lower frequency bound to f Hz. Default: 0 Hz.

-u, --freq-max=f

Set the upper frequency bound to f Hz. Default: Niquist.

-b, --fft-length=n

Set FFT length to n samples. Default: 512.

-p, --num-ceps=n

Set the number of output cepstral coefficients to n. n must be less or equal to the number of channels in the filter bank. Default: 12.

-r, --lifter=n

Set liftering parameter to n. Default: 0.

-e, --energy

Add log-energy to the feature vector.

-s, --scale-energy=f

Set scale energy factor. The way the maximum energy value is computed depends on whether `--segment-length' is specified or not.

-Z, --cms

Cepstral mean subtraction. Default: no.

-R, --normalize

Variance normalization (requires `--cms'). Default: no.

-L, --segment-length=n

Set normalization and energy scaling segment length.

-D, --delta

Add first order derivatives to the feature vector.

-A, --acceleration

Add second order derivatives to the feature vector. Requires `--delta'.

-N, --no-static-energy

Remove static log-energy from feature vector (requires `--delta').

-v, --verbose

Turn on verbose mode

-h, --help

Print a help message for the tool and exit.

-V, --version

Print version information and exit.

5.3 slpc

Usage		Syntax
Synopsis		Brief description
Options

Usage

Synopsis

Variable resolution AR modeling of the input signal.

Options

-F, --format=str

Specify the input waveform file format. Available formats are `PCM16', `wave' or `sphere'. Default: `PCM16'.

-f, --sample-rate=f

Set input waveform sample rate to f Hz for `PCM16' waveform files. Default: 8 kHz.

-x, --channel=n

Set the channel to consider for feature extraction. Default: 1.

-B, --swap

Swap the input waveform samples.

-I, --input-bufsize=n

Set the input buffer size to n kbytes. Default: 10 Mb.

-O, --output-bufsize=n

Set the output buffer size to n kbytes. Default: 10 Mb.

-k, --pre-emphasis=f

Set the pre-emphasis coefficient to f. Default: 0.95.

-l --length=f

Set the analysis frame length to f ms. Default: 20.0 ms.

-d, --shift=f

Set the interval between two consecutive frames to f ms. Default: 10.0 ms.

-w, --window=str

Specify the waveform weighting window. Available windows are `Hamming', `Hanning', `Blackman' or `none'. Default: `Hamming'.

-n, --order=n

Set the prediction order. Default: 12.

-a, --alpha=f

Set the bilinear frequency warping factor to f. Default: 0.

-r, --parcor

Output reflection coefficients rather than linear prediction coefficients. Default: LPC.

-g, --lar

Output log area ratios rather than linear prediction coefficients. Default: LPC.

-p, --lsp

Output line spectrum pairs rather than linear prediction coefficients. Default: LPC.

-e, --energy

Add log-energy to the feature vector, where the energy is the LPC filter gain.

-s, --scale-energy=f

Set scale energy factor. The way the maximum energy value is computed depends on whether `--segment-length' is specified or not.

-v, --verbose

Turn on verbose mode

-h, --help

Print a help message for the tool and exit.

-V, --version

Print version information and exit.

5.4 slpcep

Usage		Syntax
Synopsis		Brief description
Options

Usage

slpcep [options] ifile ofile

Synopsis

Linear prediction based cepstral analysis of the input signal.

Options

-F, --format=str

Specify the input waveform file format. Available formats are `PCM16', `wave' or `sphere'. Default: `PCM16'.

-f, --sample-rate=f

Set input waveform sample rate to f Hz for `PCM16' waveform files. Default: 8 kHz.

-x, --channel=n

Set the channel to consider for feature extraction. Default: 1.

-B, --swap

Swap the input waveform samples.

-I, --input-bufsize=n

Set the input buffer size to n kbytes. Default: 10 Mb.

-O, --output-bufsize=n

Set the output buffer size to n kbytes. Default: 10 Mb.

-k, --pre-emphasis=f

Set the pre-emphasis coefficient to f. Default: 0.95.

-l --length=f

Set the analysis frame length to f ms. Default: 20.0 ms.

-d, --shift=f

Set the interval between two consecutive frames to f ms. Default: 10.0 ms.

-w, --window=str

Specify the waveform weighting window. Available windows are `Hamming', `Hanning', `Blackman' or `none'. Default: `Hamming'.

-n, --order=n

Set the prediction order. Default: 12.

-a, --alpha=f

Set the bilinear frequency warping factor to f. Default: 0.

-p, --num-ceps=n

Set the number of output cepstral coefficients to n. n must be less or equal to the number of channels in the filter bank. Default: 12.

-r, --lifter=n

Set liftering parameter to n. Default: 0.

-e, --energy

Add log-energy to the feature vector.

-s, --scale-energy=f

Set scale energy factor. The way the maximum energy value is computed depends on whether `--segment-length' is specified or not.

-Z, --cms

Cepstral mean subtraction. Default: no.

-R, --normalize

Variance normalization (requires `--cms'). Default: no.

-L, --segment-length=n

Set normalization and energy scaling segment length.

-D, --delta

Add first order derivatives to the feature vector.

-A, --acceleration

Add second order derivatives to the feature vector. Requires `--delta'.

-N, --no-static-energy

Remove static log-energy from feature vector (requires `--delta').

-v, --verbose

Turn on verbose mode

-h, --help

Print a help message for the tool and exit.

-V, --version

Print version information and exit.

5.5 scopy

Usage		Syntax
Synopsis		Brief description
Options

Usage

scopy [options] ifile ofile

Synopsis

Copy input file to output file making necessary conversions. Possible conversions are normalization, dynamic features, scaling, linear transformation and component extraction.

Options

-c, --compatibility

Turn on compatibility and set the input file format to former SPro format. Default is SPro 4.0 format.

-I, --bufsize=n

Set the I/O buffer size in kbytes. Default is 10 Mbytes. If `--compatibility' is specified, the specified buffer size applies only to the output buffer, the entire input data being loaded into memory.

-i, --info

Print stream information.

-z, --suppress

Suppress data output. If this option is turned on, no output is created. This option is provided mainly for use with `--info' in order to print the stream description flag or for diagnosis purposes.

-B, --swap

Swap byte order before writing new file. Byte swapping is only possible if the output format is either HTK or Sirocco (see `--output-format' below). Default is to use the machine's natural byte-order.

-o, --output-format=str

Set the output format, where str is one of ascii, htk or sirocco. Default is the native SPro format.

-R, --normalize

Variance normalization (requires `--cms'). Default: no.

-L, --segment-length=n

Set normalization and energy scaling segment length.

-D, --delta

Add first order derivatives to the feature vector.

-A, --acceleration

Add second order derivatives to the feature vector. Requires `--delta'.

-N, --no-static-energy

Remove static log-energy from feature vector (requires `--delta').

-m, --scale=f

Scale features, multiplying them by the scaling factor f.

-t, --transform=str

Apply the linear transformation whose matrix is specified in file str.

-x, --extract=str

Extract the specified components of the feature vector. The argument str is a comma separated list of components to extract, where the components are specified either as a single index or a range of indices specified using a dash (`-'). The index of the first component is 1.

-s, --start=n

Start copying frames at frame index n. Frame numbers start with zero. Default is 0.

-e, --end=n

End copying at frame index n (included). Frame numbers start with zero. Default is to copy to the end of stream.