SPExL is the STx application for transcription (audio-to-text transcription). It is equally suited to support transcriptions and annotations of audio signals, optionally according to user controlled transcription templates. Sound file segments are created and automatically linked to *.wav files.
STx provides all methods necessary for computation of Mel Frequency CepstralCoefficients (MFCC). All the methods are described in the STx Programmer Guide. This document shows how to use the STx Macro Language to extract MFCC feature vectors from a sound file. All macros and sound processing units (SPUs) are written in the file mfcc extraction.sts which is listed in Chapter 3. This macro file has to be run from inside STx. Note that you may have to adjust the path of the #soundfile variable which has to point to a valid WAV sound file.
Read More (PDF)
STx Version: 3.7
Author: Timo Becker
Contact: Timo Becker, c/o Department of Speaker Identification and Audio Analysis, BKA (Federal Criminal Police) KT 54, 65173 Wiesbaden, Germany
A simple timer application which shows the elapsed time in a dialog and can be controlled via buttons or shell messages.
STx Version: 3.8 beta Revision 1853
Extract speech components from an existing signal and resynthesise the signal. This script demonstrates the use of the Playwave class, the SPUnits window, csigin, lpcoefs and in2table. The script takes the following parameters:
[ [ soundfile_path ] ; [ segment=Signal.All ] ; [ frame_length=20 ] ; [ frame_shift=10 ] ; [ number_of_coeficients=16 ] ]
If parameters are missing, the default values are used. There is no default soundfile_path; if no soundfile was specified, the user is prompted via a dialog to choose a file from the DataSet.
STx Version: 3.7
Author: Jonnie White
13. - 14. September 2019
ARI guest talk by Michael R. Lomnitz
19. September 2019
Seminar Room, Wohllebengasse 12-14 / Ground FloorRead more ...
27. June 2019
The Acoustics Research Institute and a number of other institutes of the Austrian Academy of Sciences want to close the gap between training in machine learning and modern data science and are ...Read more ...
20. June 2019
Kernsätze in der Mathematik besagen, dass ‚vernünftige‘ Systeme (d.h. Operatoren) als Integrale dargestellt werden können, ähnlich der Matrizendarstellung. Das erlaubt einen deutlich einfacheren...Read more ...