Hi Robert!
I'm puzzled. To my understanding, the RMS method should arrive at one value
representing an average over time. So, there shouldn't be a maximum or minimum.
Am I missing something here?
Regarding the stereo track, I suppose the average is calculated across both
channels. However, though I'm familiar with the usage of RMS, I know nothing of
how it would be coded.
Cheers! Robbie
-----Original Message-----
From: audacity4blind-bounce@xxxxxxxxxxxxx
[mailto:audacity4blind-bounce@xxxxxxxxxxxxx] On Behalf Of Robert Hänggi
Sent: Monday, February 13, 2017 4:40 PM
To: audacity4blind@xxxxxxxxxxxxx
Subject: [audacity4blind] Re: Plugin testers wanted
Hi Robbie
A quick question:
For stereo files, which RMS values should be reported?
Left
Right
Maximum
minimum
Sum
I think the maximum is adequate (same as the peak).
One can always split the track temporarily to get the individual numbers.
The sum is a bit more difficult but it is also not really important for most
users.
Cheers
RJ
2017-02-13 16:05 GMT+01:00, Robbie <tickleberryfun@xxxxxxxxx>:
Hi Robert!
I think dB would suffice. We get spoken at so much...
Shame about RMS. The one thing I miss from my SoundForge days is its
RMS Normalization feature. That was a really good thing.
I'm looking forward to further additions to the suite.
Cheers! Robbie
-----Original Message-----
From: audacity4blind-bounce@xxxxxxxxxxxxx
[mailto:audacity4blind-bounce@xxxxxxxxxxxxx] On Behalf Of Robert ;
Hänggi
Sent: Sunday, February 12, 2017 9:26 PM
To: audacity4blind@xxxxxxxxxxxxx
Subject: [audacity4blind] Re: Plugin testers wanted
Hi Robbie
Thanks for your kind response.
Comments inline.
2017-02-12 17:21 GMT+01:00, Robbie <tickleberryfun@xxxxxxxxx>:
Hi Robert!
Sorry I'm a bit late. I didn't get around to testing it sooner.
I like the idea. To me the speech rate and volume are OK, but that of
course is entirely down to personal preference.
A couple of remarks. under "Add/Remove Plug-ins" the plug-in appears
as "tts-suite" if not yet enabled. Once enabled it appears as "Report
Peak".
Yes, I probably have to adjust this. First, I thought it would be wise
to have the same prefix for the sound and code files and to keep them short.
However, this shouldn't be a problem for Windows only.
I'd be very keen to have the RMS level. That would be a great help.
Elapsed time and Selection start would also come in handy. To hear
them at the press of a key would ease the workflow.
There are some values that are "instantaneously" available.
- Peak
- Gain
- Pan
- Selection start
- Selection end
- Frequencies in the spectral selection toolbar
- Number of clips
- Mono/Stereo
- Rate
- Format
and some more
RMS does at the moment not belong in this group and neither does pitch
or tempo.
They will need some processing time. I'm not sure how long exactly.
I've asked Steve to enable access to the cached RMS values from within
Nyquist.
For now, including the next release, we have to put up with the
calculated version.
I would prefer not to have the type of value, Peak, RMS, etc.,I fully agree. I had "Peak level" at the beginning but it is rather
spoken, because I know my shortcuts. But I suppose that also is a
matter of personal preference.
disturbing and distracting information, especially if you listen to
more than one track report in succession.
(an exception might be if we provide peak and RMS at the same time) Is
Decibel" OK or do you prefer "DB"?
The latter is a bit shorter as well.
In general, the unit should be included, I think.
See also below (dB, linear, %).
Of course, there is the language issue. I think the voice fileEach phrase needs its own start and end times.
shouldn't be that much of a problem, but how easily can the code be
adjusted? Is it a matter of "Find and Replace"<?
There are two possibilities:
1. A voice file that has each phrase in a constantly long frame,
about 1 second apart. A new language would start at a constant offset,
e.g. at second 120 comes French.
2. The phrase lengths are minimized. Each language had therefore its
own set of start and end times.
The first version has the advantage that the start and end times
haven't to be explicitly defined, The list of phrases in the correct
order would be enough. However, the phrases have to be shorten
afterwards (removing digital silence/white space). The file size is probably
also about twice as big.
What's more, it's very time consuming to make such a file.
I prefer the second version. It is easy to just append other languages
and time maps.
I'm considering a plug-in that is used to generate the other plug-ins.
It would have the following controls:
- Text box: Name of the report plug-in
- multi choice: language (if available)
- Multi choice: report to add to the plug-in (one per execution) For
instance:
reset (deleting all reports)
peak (dB)
peak (linear)
peak (%)
and so on.
The preview would naturally play the current report choice.
One can have more than one value, e.g. peak and RMS, or gain and pan.
However, this will probably take some time to function properly and
for now it will suffice to have a collection of report plug-ins in the zip
archive.
The user can pick what he wants. After all, those text files are so
short...
Theoretically, it would be possible to integrate the list of times in
the wav file header.
Lots of possibilities.
Cheers
Robert
Keep up the good work man!
Cheers! Robbie
-----Original Message-----
From: audacity4blind-bounce@xxxxxxxxxxxxx
[mailto:audacity4blind-bounce@xxxxxxxxxxxxx] On Behalf Of Robert ;
Hänggi
Sent: Tuesday, February 7, 2017 10:21 AM
To: audacity4blind <audacity4blind@xxxxxxxxxxxxx>
Subject: [audacity4blind] Plugin testers wanted
Hello Fellow-listers
I'm working on a plug-in suite that should facilitate working with
Audacity.
The goal is to have a collection of self-voicing plug-ins that report
commonly wanted information.
*** works only on Windows ***
The first one announces the peak level of the selection from -120 dB
up to
+120 dB.
I'm attaching it along with the needed voice file.
Also available at
https://www.dropbox.com/s/keze2sx5lv90pxg/tts-suite.zip?dl=0
Unzip the two files into the plug-in folder of Audacity.
under "Add/Remove Plug-ins" search for "Report Peak"
It will appear in the Analyze menu if enabled.
I highly recommend that you assign a shortcut to it in order to work
more efficiently with it.
Features:
- self-voicing
- Range from -120.0 dB to +120.0 dB.
- says "less/more than..." if the audio lies outside the range.
- says even numbers (zero after the point) as e.g. -12 dB
- adds one digit if need be, e.g. -76.3 dB
- works with multiple tracks selected
- no need to press OK or something like that
I'd like to know:
- how about the speech rate, too fast/slow?
You can tweak it a bit by opening "tts-peak.ny" in a text editor and
changing the "1.0" in the very last line to something different (e.g.
0.9 will be slower and deeper)
- How about the volume?
It's also adjustable in the last line; it's the -6 which means half
volume.
- Do you want a different voice?
I can give you the text that is needed for this and future plug-ins.
Send me the audio file back or I'll explain how it should be done.
- Do you want "Peak" or "Peak Level" spoken as well?
They are included in the voice file and it needs just adding them
after "db"
in the plug-in file, e.g. "db" "peak" "level".
- What other self-voicing info do you like to have?
The voice file is already prepared to speak other things.
- numbers up to 1 million (excl.)
- Note names (pitches) including flat sharp and cents.
- RMS and peak level
- bit depth
- Hertz
- stereo, dual mono and mono
Are there other things that come to your mind (e.g. time percent
etc.)
Big thanks to those that are going to test it.
Looking forward to your criticism. ;-)
Best
Robert
The audacity4blind web site is at
//www.freelists.org/webpage/audacity4blind
Subscribe and unsubscribe information, message archives, Audacity
keyboard commands, and more...
To unsubscribe from audacity4blind, send an email to
audacity4blind-request@xxxxxxxxxxxxx
with subject line
unsubscribe
The audacity4blind web site is at
//www.freelists.org/webpage/audacity4blind
Subscribe and unsubscribe information, message archives, Audacity
keyboard commands, and more...
To unsubscribe from audacity4blind, send an email to
audacity4blind-request@xxxxxxxxxxxxx
with subject line
unsubscribe
The audacity4blind web site is at
//www.freelists.org/webpage/audacity4blind
Subscribe and unsubscribe information, message archives, Audacity
keyboard commands, and more...
To unsubscribe from audacity4blind, send an email to
audacity4blind-request@xxxxxxxxxxxxx
with subject line
unsubscribe
