REW can export the measured impulse response directly as MLSSA formatted text file, which can be used within the VituixCAD IR -> FR tool, however the way REW exports this data is a bit weird. By default, it provides 1 second of data prior to the peak of the impulse, which is not very useful for retaining the timing information, t=0 on the impulse response chart. One way around this is to apply windowing to the impulse response file prior to export.
The following procedure will erase all data prior to t=0 in the measurement, so make sure that the recorded impulse occurs after t=0 for this process. The measurement guide for REW for VituixCAD mentions to apply a timing offset when measuring, so the impulse occurs at t=0. This is due to REW's odd way of processing phase, which uses t=0 as the timing reference for phase, rather than the window reference time location. Applying the timing offset removes the excess phase from the measurement for more readable phase information.
For this process I would suggest not using a timing offset, in which case t=0 will be the loopback reference time, and the impulse will occur a few ms after t=0.
To export impulse response from REW with timing information included, set the IR window to 0ms (t=0), left window to 0ms, and set the right window to 1000ms to capture the entire measurement data. This will ensure that the first sample in the exported data is t=0, and the exported data is effectively not windowed. Note that if the impulse data occurs prior to t=0, like in a single channel measurement where impulse peak is at t=0, the window reference time should be decreased to a negative value to capture the impulse data. I would suggest 2ms prior to the impulse.
Export the impulse response as text, un-check “normalize samples to peak value” to maintain the measured amplitude information, and check “Apply IR Window before Export” to export only the windowed information.
That's it, you can now process the impulse responses in VituixCAD following the guide for ARTA, starting at page 7 "Exporting far field responses".
https://kimmosaunisto.net/Software/V...ement_ARTA.pdf
0
0.020833
262144
3.276819E-07
2.736350E-07
...
1st line is just 0
2nd line is sampling interval in ms
3rd line is length of data
Data starts at line 4
So REW's text export is not MLSSA format.
One possible solution is that I modify text file reader in Convert IR to detect sample rate, start time and length from REW text format, and skip most of empty samples (with negative time) automatically. After that process would be closer to ARTA and CLIO.
Sorry, I’m not well versed in MLSSA file formatting. The problem observed if I export the impulse from REW directly without windowing is that the peak of the impulse occurs always at 1 second mark, so timing information of t=0 is lost in the export process. Applying the window allows for sample 1 of saved impulse file to be at t=0.