Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - Robert Fraser

Pages: [1] 2 3 ... 7
General Board / Re: Problem running Soundshaper on Windows 11
« on: May 06, 2024, 05:20:20 PM »
When you mentioned the "SETUP PROBLEM" message, it's a pity you didn't list the apparently missing folders. The message means that the value (folder-name) assigned to each of these listed folders is invalid - the folder does not exist.  Several of them are read in from the textfile Soundshaper.cfg - I'm wondering if your student copied an old Soundshaper.cfg file from somewhere and the program is looking for the wrong folders? (It's impossible to tell without the list of missing directories.)  However the one that could be faulty here is actually a fixed one: namely Soundshaper's \BAT sub-folder. If that were wrong for any reason, Soundshaper couldn't read the CmdData file and therefore the parameter pages would show junk instead of proper labels or ranges, exactly as you report.

General Board / Re: Problem running Soundshaper on Windows 11
« on: May 04, 2024, 05:49:28 PM »
"Soundshaper works very fine on WINESKIN port for MAC"
This could be very important for MAC users - I looked up Wineskin, but found different versions and I'm not sure which to download. Any help you can give on this, and how to use etc., would be hugely appreciated. 

General Board / Re: how to get txt for FOFS?
« on: May 04, 2024, 05:43:37 PM »
In the bash script above, .aiff should ideally be .wav nowadays (infilepchdummy.wav) and CDP is also advocating the use of PVOCEX instead of .ana for freq. analysis files  - i.e. "infile.pvx" instead of "infile.ana".  Pvx ensures compatibility with CSOUND; .ana still works though and is used in Soundshaper, which doesn't support .pvx yet.

For the PC version, check that you have the replacement REPITCH.exe (see post under the "Announce" category).

General Board / Re: how to get txt for FOFS?
« on: May 04, 2024, 11:07:37 AM »
The PSOW (FOF) programs require a pitch-trace breakpoint file extracted using PTOBRK WITHZEROS.  And that requires a pitch file (.frq) extracted from a frequency analysis file (.ana), using REPITCH GETPITCH 1.  This is all better put in a little script (Windows CMD batch file or MacOS bash shell script).

Here are two similar scripts: the first kindly supplied by Richard Dobson for bash, and my own batch file (PC) based on Soundshaper. Both make use of arguments supplied on the commandline - particularly the infile and outfile names. Bash scripts have variables $1, $2 and so on, referring to the first, second etc. arguments; the equivalent in CMD batch files is %1, %2, etc.   Both scripts assume a mono input sound.

First, Richard's bash script frqtotext: note the usage "frqtotxt insndfile outtextfile", where outtextfile is the name of the pitch-trace file.

echo - shell script to prepare pitch-brkpnt-data file for
echo                use with a FOF-source file: lengths must match.
if [ $# -ne 2 ]
        echo usage: frqtotxt insndfile outtextfile
        exit 1
echo pvoc anal 1 $1 infile.ana
pvoc anal 1 $1 infile.ana
echo repitch getpitch 1 infile.ana infilepchdummy.aiff infile.frq
repitch getpitch 1 infile.ana infilepchdummy.aiff infile.frq
echo ptobrk withzeros infile.frq $2 20
ptobrk withzeros infile.frq $2 20
echo Delete \(temporary\) files no longer needed:
echo rm infile.ana
rm infile.ana
echo rm infilepchdummy.aiff
rm infilepchdummy.aiff
echo rm infile.frq
rm infile.frq

Now the BAT version, which adds parameter values for the application PSOW STRETCH. In this example the file suffixes are in the script, so would not be supplied on the commandline:

REM %1 is input wav (just the name without .wav); %2 is the output wav (also without .wav)
REM %3 is amount of stretch; %4 is no. of grains
echo off
REM  Convert wav to ana
pvoc anal 1 %1.wav %1.ana -c1024 -o3
REM Extract pitch
repitch getpitch 1 %1.ana %1pch.ana %1.frq 
REM erase unwanted pitch file             
erase %1pch.ana   
REM Create pitch-trace text file                                   
ptobrk withzeros %1.frq %1.pch 20   
REM Stretch the input sound                     
psow stretch %1.wav %2.wav %1.pch %3 %4   
REM Erase intermediate files...             
if exist %1.ana erase %1.ana                             
if exist %1.frq erase %1.frq                             
if exist %1.brk erase %1.brk
REM Optionally play the output
if exist %2.wav paplay -i %2.wav 

A typical commandline for this might be fofstretch my mystr 2.0 20. For test purposes, put the script in the same folder as the input file.

General Board / Re: Problem running Soundshaper on Windows 11
« on: May 03, 2024, 05:19:48 PM »
I've done a fresh installation on my wife's Windows 11 laptop. So far, everything seems to run as it should (except for F keys which are presumably set to Windows things).  I did have to get past Windows and the security software not wanting me to run the installation program, but hacked round this.

So far I've tested with modify speed, followed by repitch transpose (to test spectral auto-conversion).

Because of the vast number of CDP processes, Soundshaper reads parameter data from a text file CmdData.txt which must reside in its \BAT sub-folder. Any issue with reading the appropriate setup data from this file will give strange results and messages as you report. So that's the problem - the file is easily replaced if corrupted somehow...  Can your user test first if the file is there, in the correct place?  Have they installed Soundshaper/CDP from fresh or just copied files from someone, which might mean the setup is not right?

It may be necessary to correspond directly to get to the bottom of the issue. Your student is welcome to send me emails in Italian and I can use Google Translate to understand and reply. Email is found here:

Robert F.

General Board / Re: Problem running Soundshaper on Windows 11
« on: May 03, 2024, 03:52:29 PM »
Thanks for your bug report. The parameter pages are clearly not loading for some reason. Not sure at this stage whether this is genuinely a Windows 11 problem or not. I'll test the program on Win11 and report back.

Announce / Replacement file for faulty REPITCH program (PC only)
« on: March 27, 2024, 10:05:32 PM »
We've found a fault with the PC version of the program REPITCH, affecting the important GETPITCH function. The version originally distributed with CDP Release 8 (dated 15th March '23) is not working. This also affects further functions which depend on pitch extraction.

An earlier compilation, dated 2 Jan. 2023, appears to work as intended, however. You can download it here:
In your CDP programs folder, simply replace the 15th March '23 version of repitch.exe (if that is what you have) with the January one.

For new users, the PC program set for Release 8 will be updated tomorrow with the working version.

General Board / Re: how to create hybrids of 2 sounds using formants?
« on: March 21, 2024, 09:13:41 PM »
I've written a tutorial on how to create hybrid files in Soundshaper. There are many screenshots but no sound examples. This can be downloaded here:

Towards the end of writing it, when dealing with extracting pitch and formants, I found that the CDP function repitch getpitch was not working with the March'23 compilation (the one distributed with Release 8 ). An earlier version compiled in January '23 does seem to work, so I've made this available as a temporary fix while we look into the fault. You can download it here:
In your CDP programs folder, simply replace the March '23 version (if that is what you have) with the January one.

Robert F.

General Board / Re: how to create hybrids of 2 sounds using formants?
« on: March 20, 2024, 10:21:52 PM »
CDP has many processes requiring two input files. In Soundshaper these are normally loaded into two processing rows - A and B. Row A is selected by default but if necessary, click on Row A, then import a file by drag+drop or file menu, etc. into cell A_0. Then similarly click on row B and import the second file into cell B_0. Re-click Row A if this will be the primary source.

Select your process using the menus. A red box appears on the left side for selecting the 2nd source. Simply click on cell B_0. Then click OK in the box. If you are using spectral processes (for formant applications), both files should be converted automatically to spectral .ana files, behind the scenes. Select parameter values and click RENDER. The result should now be in cell A_1 - the next in the A row.

A simple example is CROSS, similar to VOCODE, which replaces the spectral envelope of the A sound with that of the B one. If the A is more static and the B rhythmic, you get a good result.

A word of caution: CDP expects the two files to be of compatible type. Processes won't work if one is 24-bit and the other 16-bit, for example.  If not sure, click on the cell and select CTRL+I for file information. All CDP spectral processes are natively mono-only, but Soundshaper can handle stereo files. I tested my CROSS example with a mono sound B modulating a stereo sound A and this worked, but I can't guarantee this for all CDP processes, unfortunately.

In Soundshaper's Patch grid, each row is a chain of processes from left to right, and the last cell in a row is the latest one for that chain, ready for the next process. When using two or more inputs, the 1st (or main) input will always be the last in its row, but the secondary ones can come from any cell, even one earlier in the same row. For example, if the source is in A_0 but the processed output in A_1 is significantly quieter, you might select LEVEL | BALANCE then click A_0 as the second input. This would balance the original and the processed sound.

Further information can be found in "Soundshaper Basic Operation", a pdf available from the "Graphic" page in the CDP online documentation, which you can access from the CDP website.

Mac installation / Re: Going to use the terminal in OSX
« on: October 02, 2023, 02:06:28 PM »
You're right, Frank, the old email is gone, at least for now. (I posted a notice about this I think.) My web-host wanted money for each email address, formerly free. Please use ensemble-software at . 
Thanks, Robert     

PC Installation / Re: Missing Folder - InvertPch?
« on: October 02, 2023, 01:52:10 PM »
I've posted a general follow-up called "Data Files in Soundshaper". A little more on InvertPch, though:

There appears to be a bug when processing stereo files. The script will need revision and I'll look into that.
The original CDP InvertPch works with extracted pitch-data files (.frq). These are then combined with a formant file (.for) for re-synthesis.
  Soundshaper's "Invert pitch contour" (not the "Alternative method") does this all within one script, extracting and altering the pitch data and then resynthesising using a formant file taken from the original sound. (The lower-level method is still available via the PitchData menu.)
I ran a test using a simple speech file of four words, three of which are clearly around MIDI 55 (tenor G3) while number 3 dips down a semitone.
  I used a map file of -1 11 and 1 -11 and an ABOUT value of 55 (ie invert around G3) and got a fairly convincing inversion of the sound an octave higher with the semitone shift going up instead of down (G4 and Ab4).
A different ABOUT value (60) gave a similar result transposed up to E5/F5: I'm not sure I understand the logic of the inversion here (I didn't write the program!).
The "Alternative method" uses the Release 8 (forthcoming) SPECFNU 15. This processes formant values directly. To my ears this would be less useful - more muddy.
Overall, playing with formant and/or pitch can only degrade the original sound quality in some way. For example, altering pitch only "works" to the extent that the new frequencies are also present in the formant file it's combined with. (For this purpose, you might consider transposing the original sound and extracting that formant file.)
I tried extracting the source's pitch as a text file and displaying that in the Graph Editor, the idea being that the display might tell you the pitch content. It probably does, but it looked over-detailed and I trusted my ear and a nearby keyboard better!

General Board / Data files in Soundshaper
« on: October 02, 2023, 01:48:48 PM »
This post is in response to a query about datafiles in Soundshaper (see "Missing Folder - InvertPch?" , Oct 1 2023).

Not being a realtime system, CDP works with (and is dominated by) files: source sound files (or other types), processed files to save, automation via breakpoint text files, and more specialised data files most of which are described in the documentation (see the link to File Formats from any reference Program Group, e.g. BLUR).

When a data file option is selected in Soundshaper's parameter page, a text box appears for the filename (instead of a control for a fixed value). This is filled in with a default value of the Process Name and a Filename; the Process Name is given as a subfolder of the user's Text Data folder. If an actual file exists, it is displayed as text in the Data Edit Panel (to the right of the page). Many parameters have such files - usually very basic templates - especially breakpoints, mainly so that these can be edited in the Graph Editor.

For more specialist data, a file might not always be supplied, so the Data Edit box is empty (for the user to fill in) and the Process Name subfolder may not exist either. This may seem like an anomaly, but it allows you to fill in suitable data and click SAVE to save this to the default location from where it will be recalled automatically next time. Having a standard folder for each process also allows you to save your own datafiles there for easy recall, although an alternative is to save these to a project folder of your choice.

When datafiles are amended, the SAVE CHANGES button is activated to save the data to a temporary file. These temp files are an important aspect of Sounshaper's process chains (shown on the patch grid); to re-edit or re-run the process it must have a specific datafile for that particular operation. (You might have used the same process with different values elsewhere in the patch.) So don't be surprised if your default file changes from e.g. \MagicProcess\scrunch.brk to something like ~A_1\scrunch.brk, or ~X_1\scrunch.brk for a preview.

When patches or presets are saved, the temp files actually used are saved with the patch or preset, for future recall.

Mac installation / Re: Going to use the terminal in OSX
« on: October 02, 2023, 10:58:14 AM »
I'm afraid life (preparing for CDP8) got in the way here too! No further progress on updating the Creative Sound Design tutorials, I'm afraid, but I should be able to complete this task from late October. I can post a notice to the Forum when done.

PC Installation / Re: Missing Folder - InvertPch?
« on: October 01, 2023, 10:06:59 PM »
When the user selects a datafile, Soundshaper looks in a default location (folder) for an appropriate file and displays it if it exists. In the case of an Invert-Pitch Map, no such file has been supplied, so there is no missing folder.

One reason for not supplying a template file in this case is that the data would be rather specific to the sound you would want to invert and how you personally would want to warp the intervals. The file itself is easy enough to write: if, say, you wanted every semitone rise to go down 11 semitones, you would write 1 -11 (the data is paired). and if you also wanted every descending semitone to go up eleven semitones, you would enter: -1 11.

General Board / Re: How to add infile 2?
« on: August 23, 2023, 05:46:13 PM »
Hi Phoenix,
The rows in Soundshaper's patch grid are chains of processes, each going from left to right. So your source should appear in A_0, not A_1, and when you've processed it, the output will be in A_1. (A_2 would be for the output of the next process, if any, not for the second input.) For your second input, place the cursor on cell B_0 and drag your second soundfile onto the screen (or use the file selector). Click again on cell A_0 and select the process that requires a second input. A red "ADD INPUT" box should appear. Click on cell B_0 to select it and click OK in the red box. If your process is spectral, B_0.wav will be automatically converted to B_0.ana behind the scenes.

Set your parameters on the parameter page and click RENDER to return to the main page. The output should be in cell A_1. For a spectral process, this will be colour-coded aqua and should again be auto-converted back to .wav to play it. However, you can also hit F9 to do an explicit ana-wav conversion, if you like. This would be worth doing if you save the process as a patch, for example.

Hope this helps.
Robert F.

Pages: [1] 2 3 ... 7