 The Clarinet BBoard
|
Author: kdk
Date: 2021-01-02 23:43
As I think I've mentioned here before, I've been arranging a lot of duets, originally for many different soprano-alto instrumental combinations, for my violist wife and I to play, especially while all of our playing outside our home is on hold because of COVID-19.
One great source is the group of DVDs of clarinet music compiled by the Clarinet Institute of Los Angeles. There are now 4 disks, and there is lots of duet material on them.
When I try to read the .pdfs into my OCR software to begin rearranging the duets for clarinet and viola, I get an error prompt telling me the file is encrypted and I need a password to read it. I don't think any of the music on those disks is protected by copyright. I can copy the individual files to other places on my hard drives and I can print them, both without a password. Does anyone, maybe someone connected with Clarinet Institute, know if this is some kind of protection placed on the files by CI? If so, why? or what the password is to allow full personal use?
Or could this be something to do with the .pfd format they use and not deliberate encryption?
FWIW, I get the same prompt when I try to read OMCDRL files into my OCR program.
Karl
|
|
Reply To Message
|
|
Author: Matt74
Date: 2021-01-03 02:34
Maybe I can help get you on your way. I don't have Acrobat installed (grumble, grumble), so I'm going from memory. The problem with software is that it's written by computer programmers. They can't understand NOT being a programmer, any more than normal people can understand BEING one. The error message is not designed for regular people, so it tells you what a programmer would think, not a normal person. IDK for sure, but the problem is probably not encryption...
PDFs have a number of security settings. You can set them so that you can't open them without a password, or you can make it so that anyone can open them, but not edit them, etc. It sounds like you can open and view them, so the OCR should be able to read any text they contain.
OCR is "optical character recognition", it can ONLY READ TEXT, not music notation or other images. I think the problem is that your PDFs are musical notation, not regular text.
OMR (optical music recognition) is for music. Apparently they used to call this "OCR", but that was confusing for obvious reasons. There are some OMR software listed here: https://en.wikipedia.org/wiki/Optical_music_recognition
The reason it says they are encrypted might be because it's looking for text and can't find any (because it's an image). There might also be something wrong with the pdf file. It may have been created using notation software. It may be an old/new version, etc. Or there may be something missing from it that the OCR or OMR software is looking for - which means that Acrobat can still read it, but not your OCR or OMR. Sometimes this happens with complex file formats.
It may be possible to import the PDF to bitmap or jpeg, and then OMR it. It may also be possible to open the PDF in Reader, and then save it, and then try reading that copy with OMR. By saving it as a new file (especially if you fool with the options) it may re-encode the PDF to provide what the OMR needs.
Do you have OCR or OMR? Are you able to read other PDFs with it? Are you able to read other file formats with it (like Finale or something)?
- Matthew Simington
Post Edited (2021-01-03 03:27)
|
|
Reply To Message
|
|
Author: kdk
Date: 2021-01-03 07:03
Matt74 wrote:
> Maybe I can help get you on your way. I don't have Acrobat
> installed (grumble, grumble), so I'm going from memory.
Thanks for trying to help.
The problem isn't in Acrobat (Reader). It will read the PDFs just fine. The problem is with getting Neuratron PhotoScore to open them for "recognition" of the content.
> IDK for
> sure, but the problem is probably not encryption...
>
Very possibly true.
> PDFs have a number of security settings. You can set them so
> that you can't open them without a password, or you can make it
> so that anyone can open them, but not edit them, etc.
I don't need to edit the PDFs themselves. But something about them is preventing the music recognition software from opening the files to recognize the music content.
> OCR is "optical character recognition", it can ONLY READ TEXT,
> not music notation or other images. I think the problem is
> that your PDFs are musical notation, not regular text.
>
This is not a problem with the character type. PhotoScore, SmartScore and any of several other programs are meant to recognize the music notation (including notes and other musical symbols as well as any text that's included, like expression marks, tempo markings, bar numbers, etc.) and output it as Music XML (.mxl) which is a special case of XML (eXtensible Mark-up Language) scripting, which is itself a superset of HTML (HyperText Markup Language) script. It's designed to handle exactly what's in a music file, and the Music XML import facility in modern music notation editing software is specifically programmed to interpret the XML script, the same way an HTML parser creates a web page from HTML script.
> OMR (optical music recognition) is for music. Apparently they
> used to call this "OCR", but that was confusing for obvious
> reasons. There are some OMR software listed here:
> https://en.wikipedia.org/wiki/Optical_music_recognition
>
Well, for practical use, it's a distinction without a difference. Let's not quibble over terms.
> The reason it says they are encrypted might be because it's
> looking for text and can't find any (because it's an image).
No. It's looking for graphical music notation. And it isn't failing at recognition. It can't open the files to begin with.
> There might also be something wrong with the pdf file.
It would have to be all the files - several thousand - on the 4 disks. Plus (FWIW) another couple of thousand in the Orchestral Musicians' CD Repertoire Library, which I mentioned gives me the same encryption error.
> It may
> have been created using notation software.
The files are created by scanning existing (mostly old) printed music into PDF files. Of course, it's an open question whether or not the PDF format is in some way unusual, perhaps because of the scanning process used. That, I don't know.
> It may be possible to import the PDF to bitmap or jpeg, and
> then OMR it. It may also be possible to open the PDF in
> Reader, and then save it, and then try reading that copy with
> OMR.
Oddly enough, after reading it into Acrobat Reader it can't be re-saved. In fact, I loaded a few of the files into my Nuance Power PDF, which provides for some basic PDF editing (sort of Acrobat-light). The editing features were all grayed out - they were being blocked by the files in question.
> Do you have OCR or OMR?
It's software that's meant to read a music PDF, recognize the notation and convert it to any of several music formats, including Music XML (.mxl or .xml), but also .mid, .wav, .nif and a couple of other formats.
> Are you able to read other PDFs with
> it?
Yes. Music PDFs. Not plain text files. But even a file containing plain text will open. It just produces a blank document on the screen.
> Are you able to read other file formats with it (like
> Finale or something)?
No - it won't read output from Finale.
The program can do its own scans (but balks and wants a password for that, too) and it can read from most other graphic formats.
What I *am able* to do, since I can print these files without a password, is print them, using a PDF print driver, to a new PDF. Whatever the "encryption" is, it doesn't mind this workaround. It's an extra step, but it does work. I'd just like to understand what's going on that prevents my O/MCR from scanning the files or directly recognizing the content from the raw PDFs.
I have sent a message through the Clarinet Institute website asking about this. Maybe I'll hear from them directly and be able to report an explanation here myself one day soon.
Karl
kdk wrote:
> As I think I've mentioned here before, I've been arranging a
> lot of duets, originally for many different soprano-alto
> instrumental combinations, for my violist wife and I to play,
> especially while all of our playing outside our home is on hold
> because of COVID-19.
>
> One great source is the group of DVDs of clarinet music
> compiled by the Clarinet Institute of Los Angeles. There are
> now 4 disks, and there is lots of duet material on them.
>
> When I try to read the .pdfs into my OCR software to begin
> rearranging the duets for clarinet and viola, I get an error
> prompt telling me the file is encrypted and I need a password
> to read it. I don't think any of the music on those disks is
> protected by copyright. I can copy the individual files to
> other places on my hard drives and I can print them, both
> without a password. Does anyone, maybe someone connected with
> Clarinet Institute, know if this is some kind of protection
> placed on the files by CI? If so, why? or what the password is
> to allow full personal use?
>
> Or could this be something to do with the .pfd format they use
> and not deliberate encryption?
>
> FWIW, I get the same prompt when I try to read OMCDRL files
> into my OCR program.
>
> Karl
|
|
Reply To Message
|
|
Author: Ken Lagace
Date: 2021-01-03 07:31
I haven't read every line in all these posts, but another way I would try is to load it into an other program that can read it like Photoshop or Elements (a friend may have a copy if you don't), and save it as a PDF.
|
|
Reply To Message
|
|
Author: kdk
Date: 2021-01-03 07:33
OK - I have to update this.
I've been trying to use PhotoScore (which seems closely linked to Sibelius). I just tried to open one of the CI scores in SmartScore X2 (Musitek). Lo and behold, SmartScore had no trouble with it.
So the problem doesn't seem to be deliberate encryption.
So, I'm still curious about what's going on, but I now have to pose the question differently. What is it about those particular PDF files (from Clarinet Institute and OMCDRL) that is tripping PhotoScore up? I don't have the problem ever with files I've downloaded from IMSLP or music PDFs I've created myself. I guess a trip to the Neuratron help desk is now in order.
Karl
|
|
Reply To Message
|
|
Author: Matt74
Date: 2021-01-03 08:20
Karl, It sounds like there is a permissions issue. Nuance Power won’t let you edit them, so they must be protected. If they have it set so that you can’t save or edit it, and PhotoScore does that somehow in the process (even if you can’t see it doing it), it might tell you it’s encrypted, and won’t open it. Another application might work differently.
- Matthew Simington
Post Edited (2021-01-03 08:22)
|
|
Reply To Message
|
|
The Clarinet Pages
|
 |
.jpg){kind=small}
.jpg){kind=small}
.jpg){kind=small}
.jpg){kind=small}
.jpg){kind=small}
.jpg){kind=small}
