For eyes On Success we record interviews on several tracks in Sonar. Skype is
on one track and nancy and I are on two other separate tracks.
I then save Wav files and Nancy processes them in Audacity. Using Audacity,
the noise removal works as you would like it to work, i.e., you find a part of
the track where you can sample the noise and then apply it to the entire track
or just the selected part of the track.
One solution you might consider is to save the track with the noise on it in
Sonar as a Wav file. Then process that in Audacity and re-import into Sonar.
Audacity is fairly easy to use with a screen reader with many hotkeys for
navigating and performing functions. There is even an Audacity guide for using
Audacity with a READER THAT IS VERY HELPFUL. If you Google for it you can
easily find it.
Good luck.
--Pete
From: jsonar-bounce@xxxxxxxxxxxxx <jsonar-bounce@xxxxxxxxxxxxx> On Behalf Of
Yamuna Jivana dasa (Yogan)
Sent: Thursday, June 18, 2020 3:45 PM
To: jsonar@xxxxxxxxxxxxx
Subject: [jsonar] Help: SoundForge's Noise Reduction plugin as distructive
process in Sonar
Hi Folks,
Does anyone have some experience trying to use SoundForge’s Noise Reduction
plugin in Sonar as a distructive Audio FX process?
The problem I encounter is this: When I select in Sonar (F9 and F10), a
selection of audio containing the noiseprint I wish to remove, then go into the
Process menu, Audio FX list and from here choose Sony SoundForge Noise
Reduction, then if I select to capture noise print and preview the selection,
allowing SoundForge to capture the noise print, till that point all goes well.
But now from here, how to select the rest of the data on that track to process
the effect, not only to the noiseprint area but for processing over the whole
track? I found that we can’t just cancel out of the plugin, make a selection of
the full audio track, then go back in to process again, because this will cause
us to lose the noise print previously captured.
I found that we can get this to work by selecting from the start of the
noiseprint to the end of the entire track, then use the Automatic Capture
Timeout feature to change from capture to processing mode, then to process the
track that way using the first few seconds to automatically detect the noise
print and apply that over the entire track, and so process it that way. But
what if the noise print occurs best in a location not at the very beginning of
the track? Then how do we make this work?
Any advice will be appreciated. I’m doing a weekly radio show where I have to
process a cell phone recording of the show host, which has too much of a high
noise floor. Each time I have to first remove that noise floor from his cell
phone recording before it sounds more professional and decent. I can’t find
someone qualified on the discussion who also has a better recording device, so
I have to find a solution here.
Thanks Folks.
Kind regards,
Yamuna
From: Yamuna Jivana dasa (Yogan) <mailto:ykhandoo@xxxxxxxxx>
Sent: Monday, June 08, 2020 9:21 PM
To: jsonar@xxxxxxxxxxxxx <mailto:jsonar@xxxxxxxxxxxxx>
Subject: Re: My notes here: Sonar: How to remove deleted track audio and
prevent them from being included in a saved project
Deleting extraneous audio data from Sonar projects
When deleting audio from a track in Sonar, that data seems to remain in the
saved project, unless the entire track's audio data was deleted before saving
the project as a bundle.
From preliminary experiments the impression I get is that audio data which is
deleted from inbetween a length of existing audio on a track, does not seem to
be removed when saving the project as a bundle, or as a project in its own
folder with its audio data sub-folder.
I am experimenting with this, but since my audio Data folder is currently busy
with project data from various projects, I must wait for a time when I can
start with a clean folder.
I will then need to experiment as follows:
Load a project and add an audio clip onto that track, either by recording or
importing audio;
Delete some data from an audio track, making sure my deletion is only part of
an audio clip on that track;
Save the project as a bundle and make note of the file size of the CWB file;
Close the project and do a clean audio disk;
Make sure the Audio Data folder is empty;
Again open up the Bundle file we saved earlier, then examine what gets
extracted into the Audio Data folder.
If we find that the deleted audio still exists in at least one of the audio
files in the Audio Data folder, then we know that data still exists in the
saved project.
At this time, we try doing a Clean Audio folder function from the Tools menu,
and see if it finds any data sonar recognizes as not being part of a project.
We can test by playing the file using the Play button from the Clean Audio
folder dialog.
If we hear this is the data we deleted from the audio track, then this is a Aha
moment.
Now we tell Clean Audio Disk function to delete these audio files.
At this point, we save the project again, in the same format (Bundle file),
making sure to overwrite the existing file, just so that no new audio clips
somehow get implicated by duplication Etc during the colation process of the
Save function.
Now we look at the file size of the CWB file again and make note of it.
Compare the two file size notes we have. Does the second instance show a
smaller file size? If so we might be onto something here: We may have forced
Sonar to not save deleted data, by using the Clean Audio Disk function while
the project was loaded.
We can varify if this is true by first ensuring the Audio Data folder is empty,
then opening the bundle project again, then examining the contents extracted to
the Audio Data folder from the project, to see if our known deleted data is in
deed gone.
Note, we may want to also test all of this on the CWP saving function, and we
may like to do so with the "Use per-project Audio Folders" checkbox enabled in
the Options-Global dialog, on the Audio Data tab.
While my hands are tied with a busy Audio Data folder, if anyone would like to
run these tests reliably using all the above checks and balances to varify
findings, please do so and report your observations.
Some years ago I had discovered a lot about this problem and had written notes
on my findings, but I somehow cannot find those notes, anywhere! Now I'm hoping
I could find out as much as I had back then and now forgotten. Somehow I seem
to remember that my discovery back then was so thorough, that I may not trust
this plan as much as I did back then because I seem to recall that I had
discovered a much better method to do all this back then, but now it's all gone!
From: Yamuna Jivana dasa (Yogan) <mailto:ykhandoo@xxxxxxxxx>
Sent: Wednesday, May 27, 2020 10:25 PM
To: jsonar@xxxxxxxxxxxxx <mailto:jsonar@xxxxxxxxxxxxx>
Subject: My notes here: Sonar: How to convert an audio sample into a groove
sample or loop
Sonar: How to convert an audio sample into a groove sample or loop
These are my notes. I’m just sharing. I don’t claim at any measure to be an
expert on this subject. If someone knows better please do provide feedback so
we can all learn.
1. Import the audio onto a project track from the File - Import - Audio menu.
Note, sometimes when doing this, or when copying audio from one project to
another, the audio may not lign up correctly with the now marker, even though
we may have adjusted the settings correctly or selected the correct start time
in the Paste dialog. This might be a bug in Sonar 8.5.3. If this happens, we
should manually move it to align with the metronome. See Step 4 for this
procedure.
2. Go to Options, Project, Metronome tab, and ensure Playback is checked.
3. Ignoring the metronome for the moment, listen to the audio, then, using the
tripple-tap of Control-Shift-T, Tab, Spacebar to tap, Manually adjust the tempo
to an approximation of what it should be, so that Sonar closely aligns to the
tempo of the sample.
4. Using listening skills, identify the very beginning of the sample, and F10
to set Through marker there. Use Control F9 to set start marker to start of
project. We now have selected the leading unwanted audio. Select the track with
Comma key, then Control-X to cut the unwanted audio before the sample. In the
Cut dialog, make sure Delete hole is checked.
5. Align the start of the audio correctly with the part of the metronome
measure appropriate to the sample. In other words, sometimes the first beat of
the sample is not necessarily the first measure, first beat, but could be
example measure 1, beat 4, tick 60 (where the clock resolution is set to 120
PPQ) in the Options, Projects, Clock tab).
Hint: Once the entire sample is selected on the track, using the Slide function
can help to narrow down the positioning with the metronome by moving it back or
forwards by ticks. Using a minus sign before the number of ticks moves the
selection backwards by that number of ticks, while a number without the minus
sign moves it forward by the amount specified.
6. Once the sample is aligned with the metronome, listen through the selection
to ensure it does not drift from metronome synchronization. If it does, decide
if the sample is lagging or speading up, then slow down or speed up the tempo
until it fits without drifting. Note this will not work if the sample was
created by human musicians not following a tempo grid. For this we have to
create a groove quantize grid, which is a whole other discussion. Anyway, once
you have clearly determined the tempo, make a note of it so that we can later
use it in naming the final audio file with the BPM value as part of the
filename when we are saving it as a groove looped wave file. You can always
check it with Control-Shift-T while the project is still loaded in Sonar.
7. Now that the clip is nicely ligned up with the metronome, let's delete any
trailing audio, making sure it ends just before the next musical beat. Playing
back the sample, hit Spacebar just at the point where the music should stop,
then press F9 to see where the Now marker has landed. You can now type in the
nearest MBT value to the next beat location. For example, let's say the now
marker was at 4:04:115, so we can change this to 5:01:000, and press enter to
locate the Now marker there. Press F9 to set the From marker to that location
of the Now marker. Hit Control-F10 to select through end of project, and use
the Edit, Delete function to remove everything trailing, making sure to again
use the Delete Hole checkbox before hitting the Ok button out of the Delete
dialog.
8. Just to be sure, if the audio does not exactly start at project start, I
usually record a few milliseconds of audio at project start by arming the track
and hitting the R key immediately followed by Spacebar to stop. I do this
making sure no audio device was patched to that track's input, to ensure no
actual audio sound is recorded but only blank audio. This is because we want
Sonar to recognize audio from the very start to the very end of the sample
duration, so that it will create the proper groove clip.
By the way, I must mention at this point, that only at this time when we have
manually recorded audio onto the track, can we see audio events in the Events
List View. When we import audio (like we did in Step 1 earlier), it doesn't
seem to show up for some reason. Not sure why. I am assuming that after the
next step of bouncing all audio to a single clip, that the recorded and
imported audio combine into a single clip stretched across the track, and so
when we see the audio event in the Events List View at that stage, the single
clip may well be containing the imported sample as well. This is my speculation.
9. Now that the audio track is prepared for the proper start time and duration,
I usually use the Edit, Bounce to Clip function, just to make sure we have a
nice single stretch of audio in a single clip on that track.
10. Now we select the entire track again (comma key, F9, F10), then we use
Alt-Enter to bring up the clip properties of the clip we just selected.
11. On the General tab of the Clip Properties dialog, we name the clip if
desired. Make sure the start field shows 1:01:000, even if our actual audio
doesn't start there. Remember our silent passage from Step 8? It represents
this start of project for the sample recognition. Next tab to the Length field,
and varify that it fits the duration of the entire sample as per the project
MBT end time.
The "Original Time" read-only edit field must show us start of project:
00:00:00:00.
12. On the Audio Stretching tab of the Clip Properties dialog, on the Groove
Clip Sub-tab (which should be selected by default - [the other tab is the Audio
Snap tab which I have not yet understood]), Enable Looping checkbox must now be
checked. Beats in clip must contain the number of beats contained in the entire
sample. Note this field does not usually include the correct number of beats by
default, so be careful here. You can simply count them before getting into this
dialog and noting them down. You can also work this out from the General tab of
this dialog, by calculating them with the Length field there. For example, if
the Length field shows 2:03:000, then there are 2 measures + 2 beats (since it
ends at the third beat). That gives us 4 beats from the first measure and 2
from the second = 6 beats in total). Of course this is assuming the project is
set to 4/4 in the Meter/Key view. The remaining fields here beginning with
"Follow Project Pitch" should be left unchecked, since we are only
time-stretching rhythmic material I assume. Now we can Ok out of the dialog.
13. Now let's test if our work has paid off. Play back the project and see that
it still aligns properly with the metronome, does not needlessly speed up or
slow down, and starts and ends where it should on the timeline that we have
configured earlier. If it does, great, go to Step 14. If not, undo the
"Groove-loop properties" action, go back to step 10, and check carefully as we
retrace our steps. Still testing the success of our groove sample, change the
tempo of the project and see if the audio follows the new tempo correctly.
Note, even when it does follow, if we change the tempo more than 10 BPM, we
will notice artifacts in the time-stretching of the audio. If anyone knows how
to fix this, such as using the "Stretch Audio" feature and "Type of audio" from
the Fit to Time dialog, please let me know, because these things do not appear
in the Clip Properties dialog. I could not seem to use the "Fit to Time" dialog
on its own to make a sample into a groove-loop sample so that it will follow
the project tempo, and I suspect the "Type" field which includes things like
"MPEX- Solo instrument" Etc, will help stretch and shrink to time with less
artifacts if chosen appropriately for the audio material.
14. Finally, to save the sample as a groove looped Wave file sample, first
let's again make sure our track is selected with Comma key, our start and
through times are set to start and end of project with Control-F9 and
Control-F10 respectively, then we go to View, Loop Construction View, or press
Alt-2 shortcut. This view does not become active so we must Control-Tab to it.
Jaws will report the name of the view "Loop Construction", after reading the
file name, from the title bar of the child window. Once here, press the F key,
and it will bring up the File Save As dialog. Now we save the Wave file as
normal, but in the "Filename" edit combo field, let's remember to use the tempo
BPM value in the File Name to remind ourselves of what the default tempo of the
sample is. Check where we are saving (the "Save In" combo box and the "Folder
View" list view, and then save the file. Saving the file this way is supposed
to save the groove clip as a riff wave file.
Please save these notes in case I lose them. I'm decently good at taking notes
but bad at keeping them. I may ask some years later and hopefully someone would
have saved this.
