Signal & Metadata Management

to be contnued

 

The STx-FindFile Tool supports the user to find soundfiles on local and network disk drives. FindFile additionally performs a validity check of the soundfiles found. The result of the search is addet to the workspace of the STx workspace and integrated into the current STx project (DataSet).

The Recorder Setup show PPM bars pecifies the basic parameters, such as sampling frequency, mono/stereo, word length, maximum signal duration etc. in the left section of the dialog. Read more: Extended Recorder Setup for Online Signal Segmentation. 

Sound segments are defined by a SegmentName (ID), SegmentStartAddress and Segment Duration within a soundfile. The SegmentStartAddress is relative to the beginning of the soundfile, and the SegmentDuration begins at the SegmentStartAddress. Alternatively the SegmentEndAddress can be specified.

Assuming, you want to concatenate arbitrary sound segments from one or more soundfiles in consecutive order into one or more already existing soundfiles:

  • Open the sound segment list of the source file by selecting the file from the filelist of the STx-Workspace.
  • Just copy and paste (Ctrl+C/Ctrl+V) the selected sound segment(s) from the segment list into the segment list of the destination soundfile. Multiple selections from the STx segment list of a source soundfile are possible.

Note: The destination soundfile must already exist in the STx-soundfile list. Sampling rate and sample word length of source files(s) and destination soundfile(s) have to be compatible. Channels have to be specified.

  • Copied sound segments will automatically be appended at the end of the destination soundfile.

Note: Segment annotations are taken over from the segment list of the source file(s). 

For further options of signal and segment manipulation see STx Sequencer

Several software products, such as MS Power Point, accept *.wav files to be reproduced in presentations. In order to make STx sound segments available for that purpose the Export to Soundfile function is provided. Select the appropriate sound segment and open the pop up dialog by right mouse click and select Export→Export to Soundfile.

Export Metadata (Ctrl+M) saves the metadata for the selected file to disk.

Note that you can use the export.sts toolbox to export multiple segments to disk in the format <soundfile>.<segment>.wav 

A new sound file should be created before having STx perform a signal management or signal processing function that may change the original. This includes, filters, amplification, phase vocoder etc, which prompt the user for source and destination addresses and allow to replace the original as well as truncation, signal delete copy/paste operations. Currently STx provides no undo function!

In order to start a recording session in STx, proceed as follows:

  • Select STx-Recorder Setup
  • Specify record settings (sampling frequency, # of channels etc.)
  • Select preferred recorder layout. Note: auto-run RTAnalyser is only available under Windows XP and Vista due to system limitations in Windows NT and 2000.
  • Specify theTemp-Directory for the recorder soundfile location (or use default)
  • Start recording:  

record_start..... Stop recording and save recorder sound file to destination directory and desired name (*.wav) The recorded soundfile appears in the STx sound file list.

Note: The default soundfile naming uses the actual date and time, such as rec-2008.06.05-12.55.37.wav. 

Note: externally recorded (i.e., already existing) sound files can be imported into STx by using either of the following methods:

  • From the Windows Explorer, drag the respective file(s) into the STx-workspace and drop them there.
  • Use the FindFile tool of STx to import them into the STx-DataSet.
  • Use the File→Soundfile→Open option from the pop up menu.

Note: The STx-Recorder automatically creates a soundfile segment named Signal. All  in the soundfile segment list with the start address 0 samples and the duration of the recorded signal in xxxxyyy samples.

Image

STx currently supports sampling frequencies from 500 Hz to 2 MHz at word lengths of 8, 16 and 24 Bits, provided that these formats are supported by the hardware and driver of your Windows sound subsystem. Up to 64 channels (48 kHz, 24 Bits) have been recorded simultaneously by means of an 8 x 8 microphone array. A binary word length of > 24 bits, higher sampling frequencies and alternative soundfile formats are available on request. For further hints about performing digitization see the Sections STx-Recorder Setup, I/O and General Soundfile Setup (the latter are only available in the STx documentation/help).

A sound segment (audio segment) is a selected and named sequence of contiguous sound samples in a signal stream. Segments are located in sound files. Access to STx generated sound segments is provided by means of the Segment-List-Window, which opens when a sound file is selected from the sound file list of the STXDataSet. Segment lists specify segment addresses (cue in - cue out time pointers) and associated metadata (attributes and annotations). They support the management of named sound segments of continuous stored signals, whether located in the associated sound file or in an other sound file.

Upcoming Events

Explainable Models and Their Application in Music Emotion Recognition

ARI guest talk by Verena Haunschmid, Shreyan Chowdhury

16. Oktober 2019

14.30

Seminar Room, Wohllebengasse 12-14 / Ground Floor

Read more ...
 

Blind Output Matching for Domain Adaptation in Segmentation Networks

ARI guest talk by Georg Pichler

23. Oktober 2019

14.30

Seminar Room, Wohllebengasse 12-14 / Ground Floor

Read more ...
 

News