Discuss Notion Music Composition Software here.
17 posts
Page 1 of 1
This project is focused on discovering how Realivox Blue (Realitone) works for creating a virtual K-pop musical group in the style of T-ARA (a personal favorite) . . .

Realivox Blue (Realitone)

[NOTE: Like pretty much all K-pop songs, this one is "pumped"; so be careful with the listening volume. It's loud. Lots of "sparkles", Latin percussion, and synthesizer effects; and the visual production matches the music in great detail. I think it's amazing . . . ]

phpBB [video]


"In 2012, The Wall Street Journal reported that the cost of training one Korean pop idol under S.M. Entertainment averaged US$3 million."

[SOURCE: K-pop ~ Wikipedia ]


This is per artist, not for the entire group (which is six, where the others are dancers); so it's a big investment in training, and it shows . . .

BACKGROUND

Sometime in early 2012, I discovered that the Nirvana hit song "Smells Like Teen Spirit" apparently was inspired by what at the time was a popular deodorant among the so-called "Youth of Today" . . .

Image

phpBB [video]


Later that year in December, some of my relatives were visiting and a few of the teenagers appeared to be a bit distressed because they forgot to pack deodorant; so I volunteered to go to the local Walmart Supercenter and get them some deodorant . . .

Having no clue what to get, I asked them; and they said to get AXE deodorant . . .

I had never heard of this product, so I asked if there was anything more specific; and they said any scent will be fine . . .

When I got to the local Walmart Supercenter and found the AXE products, there were several scents; so I got the one that had a musical name . . .

The key bit of information is that combining the apparent popularity of AXE products among "Youth of Today" with the observation about "Teen Spirit" led to connecting a few dots and then having a bit of FUN composing a song that I call "Tastes Like Anarchy", which is based on another of the AXE products that was available at the time and had a perfect name for a pun . . .

Image

I composed and recorded the basic rhythm section in 2012 and composed some lyrics but never got around to singing the song . . .

I did the music with NOTION 3 and a few SampleTank 2 (IK Multimedia) virtual instruments, and recorded it in Digital Performer 7 (MOTU), which at the time were the current versions . . .

Skip forward a few years to 2017, and after experimenting with Realivox Blue enough to make a bit of progress, I decided to use Realivox Blue to do the singing in "Tastes Like Anarchy", although I might do some singing, too . . .

THOUGHTS

Although I have had some success with using phonetic scripting and music notation to get Blue to sing realistically with good diction, I was missing three key aspects, which specifically are (a) that phrases require using legato mode; (b) that the durations of notes sometimes need to overlap to keep the phonetic script stepping from one syllable to the next syllable in the phrase; and (c) that unless you use the "Repeat" keyswitch (B2) to repeat the same note, the notes need to change from one note to the next note when in legato mode, although at present I ensure that the notes change, since it's easier and more melodic . . .

Image

This can be done in NOTION 6, but it's not so easy; hence I did some experimenting with Digital Performer 9.5.1, but I don't like its MIDI Editor; so I tried the MIDI Editor in Studio One Professional 3.5.2 (PreSonus), and it's excellent for this purpose . . .

Image
Studio One Professional 3.5.2 MIDI Editor

I also tried the back-and-forth network strategy with Studio One and NOTION, which also is excellent . . .

Along the way, I discovered that on the Mac you don't need to disable the "Enable ReWire" option in NOTION 6, which makes this a lot easier, since due to the audio for the music from 2012 being in Digital Performer, I am using Digital Performer as the Digital Audio Workstation (DAW) application, and I use it with NOTION in ReWire sessions . . .

It's not completely interactive, but (a) it's sufficient for what I need to do and (b) even if I used Studio One for the DAW application, I would need to do it the same way (separate network and ReWire sessions, which are two distinctly different activities in this context) . . .

When I need to work on new phrases for Blue, I start NOTION first and then start Studio One, which prevents it from being a ReWire session where Studio One is the ReWire host controller, as is the case with NOTION, since Studio One only participates in ReWire as a host controller . . .

Doing it this way, NOTION and Studio One communicate separately via the network, although this is all on the same computer here in the sound isolation studio . . .

When I have the phrase the way I want it, I close both applications and then start Digital Performer, followed by starting NOTION, which then is a ReWire session where Digital Performer is the ReWire host controller and NOTION is the ReWire slave . . .

After making some adjustments to the music notation, which sometimes is necessary due to the fact that there is no one-to-one mapping of MIDI-to-music-notation, everything is ready to tailor the new vocal part to the music . . .

Done this way, the MIDI editing is straightforward due to the Studio One MIDI Editor, which is very precise; and it's easy to get the MIDI from Studio One into NOTION, which also is very easy to do . . .

Then I can work interactively with NOTION in the ReWire session with Digital Performer as the ReWire host; and since the "Enable ReWire" option always is ON, I don't need to reboot the Mac Pro . . .

Some of the consonants in the various Realivox Blue phrases are not so clear for the ends of syllables, but I can use Melodyne to make it better, where the general idea is to make some phrases that have the necessary ending consonants and then add them to the DNA of the full phrases in Melodyne, which probably sounds complicated, but it's not so complicated, although this is mostly a guess, since I have not tried it yet, but so what . . .

I can do the same thing with real singing in Melodyne, so it will work just as well with virtual singing . . .

In some respects, this is delving into phonetics and enunciation with a microscope; but if this is what has to be done to get the virtual singing the way I want it to sound, then it's fine with me; and after a while I will develop a system, which will make it easier (or at least faster) . . .

This YouTube video shows the Studio One Professional 3.5.2 MIDI Editor with the phrase "She tastes like Anarchy", which was the prototype melody for the phrase, except that I changed it later in NOTION to fit the basic rhythm section . . .

[NOTE: Some of the information at the AXE website suggests that their products taste good, which is where I got the idea for the pun. Yeah, it's totally silly, but so what . . . :P ]

phpBB [video]


This is the song with the prototype verses and choruses . . .

[NOTE: The "H" needs an overdubbed ending consonant or sibilant, which I will do with a new staff and instance of Blue, and then combine it in Melodyne. Everything is reasonably clear, but without the ending consonant or sibilant, the "H" sounds too much like "A". The phonetic script for "ANARCHY" is "eh een eh ahr see ehch wah ee", where the "i" part of "Y" is a diphthong, so it's two parts ("ae ee" is "i", and put a "w" in front and it becomes "Y"). I put the lyrics on the Twin 2 staff, because it's easier when I going back-and-forth in network mode revising the MIDI notes and phonetic scripts. It's actually text rather then lyrics, which also is easier to modify and to move around, since it's not tied to actual notes . . . ]

phpBB [video]


The general plan is to refine the melody and lyrics, which will include adding vocal harmonies and counterpoint . . .

Another interesting aspect is that the Korean written language (Hangul) is remarkably phonetic; so there's no ambiguity like in English. There are 40 characters, and each one has an unique sound, which fits nicely with the way phonetic scripting is done in Realivox Blue . . .

This will require a good bit of work, but it's a great way to make sense of Realivox Blue . . .

Lots of FUN! :)

The Surf Whammys

Sinkhorn's Dilemma: Every paradox has at least one non-trivial solution!
User avatar
by Marcato on Tue Dec 05, 2017 12:01 am
Quite a research lab you're operating in there, Surf!

I'll be following you progress as I'm interested in these numerous word building machines which have been appearing over the horizon in recent times.

Of course, I'll let you thrive in total anarchy as you continue grinding your Axe. Exploration seems to be your specialty anyway. On this side of the Celine Dion-Mickey Mouse line, I'm much more interested in latin phrases (the real thing!) and wish to experiment with gregorian choral spectaculars.

"Gratias agamus domino deo nostro", for instance, would sound at home with the Vienna Boys Choir or in some Benedictine Abbey, but not really at home if sung by your anarchist lady.

Still, any experimentation with word generating software serves a purpose as it shows the limits of the presently available programs.

Merci beaucoup
User avatar
by Surf.Whammy on Tue Dec 05, 2017 12:53 pm
Marcato wroteQuite a research lab you're operating in there, Surf!

I'll be following you progress as I'm interested in these numerous word building machines which have been appearing over the horizon in recent times.

Of course, I'll let you thrive in total anarchy as you continue grinding your Axe. Exploration seems to be your specialty anyway. On this side of the Celine Dion-Mickey Mouse line, I'm much more interested in latin phrases (the real thing!) and wish to experiment with gregorian choral spectaculars.

"Gratias agamus domino deo nostro", for instance, would sound at home with the Vienna Boys Choir or in some Benedictine Abbey, but not really at home if sung by your anarchist lady.

Still, any experimentation with word generating software serves a purpose as it shows the limits of the presently available programs.

Merci beaucoup


Glad you're enjoying it! :)

My interest in finding a virtual female vocalist goes back nearly a decade; and the primary reason is that I write science fiction radio plays and need a female voice . . .

For a while, I just read the female lines along with everything else--which is fine--but it's not so realistic . . .

Earlier, I was pondering the idea of hiring a female vocalist and voice-over artist, but that costs too much and is too difficult to manage; so I abandoned the idea . . .

There is a product I was considering for a while, but it wasn't a good match for my needs . . .

However, I think it might be a match for what you want to do . . .

Voices of Prague (Virharmonic)

Image

Voices of Prague has a phonetic scripting language with 7 vowels and 35 consonants . . .

There is a visual interface for constructing phrases and assigning them to the various voices (soprano, alto, tenor, bass) . . .

phpBB [video]


Having the visual interface for the phonetic scripting language is nice, and it's similar to the visual interface that Realivox Blue provides . . .

PHONETICS AND SHORTHAND

You need to know how to write phonetically, which for me is not particularly difficult due to exploring scientific shorthand for several decades, where the particular type of scientific shorthand is phonetic and is based on the way words actually sound . . .

For a while, I explored Gregg shorthand and even started taking a college-level course; but shorthand systems usually are based on the way its developers speak phonetically, and it soon became obvious that Gregg spoke with a lisp; hence learning Gregg shorthand required one to learn how to speak with a lisp, and this was too much for me; so I dropped the course . . .

Learning a shorthand system is the same as learning a new language, and writing shorthand at the same rate folks speak requires real-time mapping of what you hear to what you essentially translate and then write; so it needs to be intimately intuitive without needing to give it any conscious thought; and in this respect, it's more difficult than learning how to sight-sing music . . .

Years later, I found a book on scientific shorthand which is based on the way words sound rather than on the way a particular person speaks; and this is what I have been studying, as time allows . . .

Image

The book is out of print, and it never was very popular; but it's perfect for learning phonetic scripting and making sense of intuitively understanding the way words sound. which I suppose is a mater of linguistics . . .

The book was first copyrighted and printed in 1946, but there was a printing in 1992 . . .

Amazon.com has used copies of the book, so it's still available; and the used copies are rated based on condition and are reasonably priced. The condition-rating system is accurate and reliable; and it's something Amazon.com requires sellers to follow. You can get a copy for less than $10 (US) in good or better condition (includes free shipping) . . .

The shorthand is based on cursive writing and the circle, where the circle is used to map the various pen marks and strokes . . .

For example, "i" is the dot atop the cursive; "s" is the ascending part of a lower-case "s" in cursive; and "t" is the descending angled line of cursive "t" . . .

Image

The parts of the cursive characters are selected based on the motive direction of the pencil when you are writing, where the idea is that the letters flow smoothly when you write words and phrases . . .

When writing "it's", the phrase starts with the "t" which is written downward and then the "s" is added in an upward motion . . .

Technically, the "i" can be implied, but if you want to make it obvious, you can put a dot over the "t" . . .

So the pattern is "dot", downward short angled line, then upward small curve, where it's all above the line (dashed line in the image) . . .

After a while, you start doing abbreviations, since it's even faster . . .

Instead of the full "it's", you do it with only the downward angled "t" line; and connecting a few dots maps to "is" being only the "s" curve . . .

Then "It is" is just the "t" followed by an "s", which is easy to write very quickly . . .

A "d" is like "t" but the line is longer and continues below the dashed line . . .

Words are written phonetically the way they sound, and this is the part that applies to Voices of Prague and Realivox Blue . . .

VENTRILOQUISM AND JUGGLING

Learning ventriloquism is another way to discover how to make sense of phonetics, and it's easier than learning shorthand but nevertheless requires a lot of effort . . .

Vowels are easy, but some of the consonants are very difficult; and learning ventriloquism requires creating a mapping of the ways you say different parts of words, which in turn focuses you on phonetics . . .

In some respects, it's an interesting but mostly useless skill; but like juggling it has some benefits . . .

Juggling is a great way to improve your peripheral vision; and these three things (shorthand, ventriloquism, and juggling) create new neural pathways, hence have other benefits in the grand scheme of everything . . .

Lots of FUN! :)

The Surf Whammys

Sinkhorn's Dilemma: Every paradox has at least one non-trivial solution!
User avatar
by Marcato on Wed Dec 06, 2017 3:18 am
Hi Surf!
May I quote any portion of this heady yeat inspiring research material to put flesh on the bones of my own post-doctoral effort !? :-)

Always interesting! ;-)
User avatar
by Surf.Whammy on Wed Dec 06, 2017 7:28 am
Marcato wroteHi Surf!
May I quote any portion of this heady yeat inspiring research material to put flesh on the bones of my own post-doctoral effort !? :-)

Always interesting! ;-)


Works for me! :+1

The Surf Whammys

Sinkhorn's Dilemma: Every paradox has at least one non-trivial solution!
User avatar
by Surf.Whammy on Wed Dec 06, 2017 9:05 am
This is making more sense now . . .

I did a new version of "Tastes Like Anarchy" and among other things made a few revisions to the melody for the verses . . .

I also processed the verse vocals and chorus vocals separate with Melodyne (Celemony), and this made it possible to remove some low-level noise between phrases . . .

I revisited the chorus ("A-N-A-R-C-H-Y") and added a few measures of Realivox Blue singing "chuh" at B4, which I then used in Melodyne to add to the end of the "H", which makes it more distinct . . .

[NOTE: The DNA blobs highlighted in burgundy are the "chuh" consonants that are appended to the tails of the "H" phonemes. They're a half-step lower, which works nicely. There are three instances of "A-N-A-R-C-H-Y", and Melodyne interprets them as being several notes rather than just perhaps eight notes, which is interesting and is due to the way the phonemes are sung, which for legato includes the Realivox Blue engine adding extra samples to connect the phonemes, so that instead of each phoneme being independent and distinct, there are extra samples added during the transitions, which I suppose is what "legato" means (I verified this by finding the definition of "legato") . . . :reading: ]

Image
Chorus DNA ~ Melodyne (Celemony)

[NOTE: This is the music notation for the "A-N-A-R-C-H-Y" that is repeated over and over. There are three of these in the "Chorus DNA" image (see above), and they are exact copies. Each one is two measures . . . ]

Image
Music Notation for "A-N-A-R-C-H-Y" in the chorus ~ NOTION

Curiously, even though the music notation is the same for each chorus, the singing varied in subtle ways . . .

I added a "chuh" to the first "H" in the first chorus, and it was perfect; but it was not so easy to discover precisely where to add it to the following repetitions; so I copied the first "A-N-A-R-C-H-Y" and then used it to replace all the other ones . . .

This provides the clue that there is an easier way to do all this phonetic and singing stuff--especially since using echo units and other effects sufficiently randomizes (or "humanizes") what actually are exact copies of the same thing repeated over and over . . .

Once the chorus was working nicely, I added a "DJ scratch" effect using Timeless 2 (FabFilter Software Instruments) to the chorus, which adds a "record scratch" to the end of each "A-N-A-R-C-H-Y" to give it a bit of pizzazz . . .

And for the verse singing, I used Pro-C (FabFilter Software Instruments) to create two frequency filters--one that enhances the higher frequencies and stereo image and another one that restricts to a suitable frequency range with a tiny bit of mid-scooping . . .

The order is (a) stereo enhance it for improved clarity and then (b) constrain it to its own sonic space, where the "constraining" aspect removes low frequencies that are not needed; does a tiny bit of mid-scooping; and removes high frequencies over 15-KHz, which also are not needed . . .

Image
Stereo Enhancer ~ Pro-C (FabFilter Software Instruments)

Image
Low-Cut, Mid-Scoop, High-Cut Filter ~ Pro-C (FabFilter Software Instruments)

This is the new version of "Tastes Like Anarchy" . . .

[NOTE: I'm using headphones, so this is a headphone mix; but it started as a studio monitor mix . . . ]

phpBB [video]


THOUGHTS

The lyrics comprise a grand total of seven words for the verses and seven characters for the choruses (which actually is one of the seven words for the verses), which is fine for what I'm doing now, but it probably will change . . .

[NOTE: Part of what I'm doing now is determining the rules for the way the melody interacts with the phonemes and the way they are sung, where it appears that transitions from one phoneme to another phoneme are better when it's from lower notes to higher note, but other times transitions are better from higher notes to lower notes. I have not experimented with keeping the note the same from one phoneme to the next phoneme, but Realivox Blue supports this with a special keyswitch . . . ]

I'm also going to do some experiments where I run the singing for the verses through a series of cascading echo units, which is a technique John Lennon used . . .

For a while this was a bit puzzling to me, but after I did it on a few songs, what I discovered is that cascading echo units blur and alter whatever lyrics you are singing, with the result that what you hear after it's recorded are the correct lyrics, which often are different from what you were singing (or initially thought you were singing) . . .

Then you redo the singing using the revealed lyrics, and there you are . . .

[NOTE: I have found enough verification of this in the historical records to be confident this is a correct inference, and it matches my experiments . . . ]

This technique makes excellent sense when you think about it a while and then put it into proper perspective based on science and the principles of acoustic physics . . .

As I understand it, the idea is based on what happens perceptually when a musical group is performing a song in a concert as you walk from outside the concert hall to your seat in the front row . . .

Whatever your mind perceives as the lyrics when you are far away are the lyrics that cut through everything; and when you sing those lyrics--instead of whatever you were singing--it's correct and perfect . . .

Whether John Lennon knew this in an immediately conscious way is another matter, but he did it for whatever reason made intuitive sense to him, so it's all good . . .

Surf.Whammy's Lyrical Tip of the Day

"Singing through cascading echo units reveals the correct lyrics."

I'm going to experiment with harmonies and then do some counterpoint and whatever else the Aliens From Outer Space beam into my mind, which is the way it works here in the sound isolation studio . . .

Lots of FUN! :)

The Surf Whammys

Sinkhorn's Dilemma: Every paradox has at least one non-trivial solution!
User avatar
by Surf.Whammy on Sun Dec 10, 2017 7:48 pm
I did a bit more work on the Realivox Blue melody this afternoon, which included changing the melody for the ending phrase of each verse . . .

[NOTE: The Aliens From Outer Space probably know how this song is supposed to sound, but it's all new to me; so the more I listen to it, the more I understand, which mostly is "by ear". The strategy is that I listen to the current version for a while, and then a new idea or a correction appears, and I make it happen. When you do everything yourself, this is the way it works . . . :+1 ]

The first enhancement is adding the phoneme "stah" near the end of each "taste" or "tastes" in the melody to make the words more distinct . . .

[NOTE: The "stah" DNA blob is highlighted. The singing mode is legato, so it's fine for the "stah" to be a slightly lower (or higher) pitch than the "tays". It happens so rapidly that it sounds natural . . . ]

Image
"stah" added to melody DNA ~ Melodyne (Celemony)

The strategy for doing this begins with adding some quarter notes singing "stah" to a measure in the song where nothing vocally is happening, followed by removing the Melodyne (Celemony) plug-in from the Digital Performer (MOTU) track and then recording a new version of the NOTION track . . .

When Melodyne is being used, it has to be removed when you record a new track for an existing track, since once you do the audio transfer to Melodyne, it replaces the recorded audio . . .

Otherwise, you see NOTION following the notes and hear the NOTION generated audio if you are input monitoring; but after you record the track, what you hear is whatever Melodyne has as its audio, which is a bit subtle and took me a few minutes to understand, although now it makes sense . . .

[NOTE: This YouTube video shows Melodyne and the DNA "blobs" for the primary vocal melody performed by Realivox Blue via music notation and phonetic scripting. This is best enjoyed when listening with headphones, since there's motion and "sparkles" . . . ]

phpBB [video]


The next enhancement is adding a second track of Realivox Blue singing harmony to the primary melody . . .

The harmony is a chromatic fifth above the primary melody, and all I had to do was create a new staff in NOTION; copy the music notation for the primary melody and paste it to the new "harmony" staff, followed by doing a transform (chromatic fifth, above, same octave) . . .

Image

When doing this it's important to do the transform only for the treble staff, since the bass staff has the keyswitches that tell Realivox Blue which phrases to use . . .

[NOTE: I adjusted the relative volume levels of the primary melody and the harmony, and ran the harmony through Reel ADT (Waves) to "double-track" it for clarity. The harmony is somewhat subtle, since it's secondary to the primary singing. This is only on the verses, not the "A-N-A-R-C-H-Y" choruses, which are unchanged and are on a separate track. Now there are three tracks of Realivox Blue (melody and harmony verses, and the choruses). This is best enjoyed when listening with headphones, since there's motion and "sparkles". . . ]

phpBB [video]


There are a few times in the verses when "anarchy" sounds like "eh-ar-dee", "eh-ar-gee" or "in-ar-dee", so I need to work on that a bit, which I think will be easy to correct now that all this phonetic singing stuff if making more sense . . .

It appears to occur depending on whether the notes are sequenced high-to-low or low-to-high, hence I think the pitches of the notes have affect the phonemes, which probably is due to the extra samples added between notes when the singing mode is legato . . .

THOUGHTS

Conceptually, this is like producing and recording a real female soprano vocalist who needs a bit of guidance for enunciating, phrasing, and so forth . . .

I imagine myself in the control room; stopping the recording; and giving the guidance, "Let's redo that phrase, and this time over-enunciate 'taste' by adding 'stah' at the end, so instead of 'taste', sing 'tays-stah'" or perhaps doing some overdubbing to add clarity to already recorded vocal tracks, since if everything is good except for the starts or endings of a few words, there's no need to redo everything . . .

The key differences are (a) that it's all virtual and (b) that you are doing the music notation, phonetic scripting, and overdubbing . . .

It takes the same amount of time as it would take with a real female soprano vocalist, so even though it takes a while to do the music notation, phonetic scripting, Melodyne editing, and so forth, (a) it's fine with me and (b) it works . . .

Lots of FUN! :)

P. S. The Aliens From Outer Space just beamed me a new instruction: "Auto-Tune" . . . :P

The Surf Whammys

Sinkhorn's Dilemma: Every paradox has at least one non-trivial solution!
User avatar
by Surf.Whammy on Wed Dec 13, 2017 9:24 am
The Aliens From Outer Space beamed me the theme for "Tastes Like Anarchy" yesterday, and now it's a science fiction song about cyborgs . . . :P

The concept is that sometime in the future the only way to determine whether someone is human or cyborg is by tasting, so instead of shaking hands, everyone tastes hands when greeting . . .

Yeah, it's stupid, but so what!

I added a series of phrases to the chorus, which took about 18 hours . . .

The new phrase in the first chorus is "She's a cyborg", and there's a corresponding set of phrases for the other choruses as they cycle through pronouns {she, you, I, we} . . .

"A-N-A-R-C-H-Y" is now the acronym for "Artificial Neural Alpha Recombinant Chromosomal Heuristic Ytterbium", which is the unique taste common to all cyborgs . . .

[NOTE: The chemical symbol for ytterbium is "Yb", and its atomic number is 70 . . . ]

The song is making more sense, and I am starting to hear more stuff . . .

[NOTE: This is mixed for headphone listening, and there's lots of motion-based effects and stereo imaging, especially for the custom echoes done with Timeless 2 (FabFilter Software Instruments), which took as much time as the new Realivox Blue choral phrases. There's a new phrase at the end of the song, which is easier to hear, since it's almost a cappella . . . ]

phpBB [video]


In this version, there are five tracks of Realivox Blue, and four of them are additionally processed with Melodyne (Celemony(, which is the way I add overdubbed consonants to the tails of words and fine-tune the singing so it sounds as realistically "human" as possible . . .

Lots of FUN! :)

The Surf Whammys

Sinkhorn's Dilemma: Every paradox has at least one non-trivial solution!
User avatar
by Surf.Whammy on Mon Dec 25, 2017 7:40 am
In the serendipity department, I was having a conversation with a graphic artist over the weekend about an icon for my pretend Asynchronous Demi-Melodic Death Metal Nouveau musical group, KnightKocK, and I thought it would be helpful to provide some YouTube videos of the even more silly Metal musical group, GWAR . . .

[NOTE: If you like KISS, then GWAR is like KISS cranked to 11 . . . ]

I suppose it's a combination of (a) being older and (b) having a functional brain, but some of the Metal stuff is remarkably silly--hence must be parodied and mocked vigorously, which is the reason I created the pretend musical group, KnightKocK . . . :P

The serendipity aspect occurred when I noticed in the YouTube suggestion list that there was a Japanese Metal vocal group called "BABYMETAL" . . .

[NOTE: This is "pumped", so be careful with the listening volume level. The singers are excellent vocalists. I thought K-pop was a bit wacky, but I was wrong. This video was done in 2014, but more recently BABYFACE has been touring in the US in 2017 as the opening act for Korn. If this vocal group were stock, then I would buy a lot of it, because they are going to be big . . . :+1 ]

phpBB [video]


THOUGHTS

The fascinating aspects of "Gimme Chocolate" (BABYMETAL) as it applies to "Tastes Like Anarchy" are the rapid vocal phrases, which are frantic and very similar to the "Morse Code" style of the lead guitar in "Tastes Like Anarchy" . . .

There are some synthesizer vocal pad effects that I like, as well; so this is something I am thinking about exploring--perhaps using Volcano (FabFilter Software Instruments), which has some vocoder presets, although for live performances they're probably using something like one of the TC-Helicon vocal processors . . .

In the new version of "Tastes Like Anarchy", I added a second primary melody track and spread the three primary melody tracks, where one of them is an echo track; and I added a "La-La-La' vocal phrase at the end of each chorus . . .

The new double-tracked primary melody track is run through Timeless 2 (Fabfilter Software Instruments) for customized "Plate" reverb-style echoes; and the "La-La-La" stereo track has custom echoes, which also were done with Timeless 2 . . .

[NOTE: Adding a separate double-tracked Realivox Blue voice track and adding a cascaded-echo track for the primary melody (verses) to create vocal spread is a bit subtle, but it works . . . ]

For reference, one of the reasons I process the singing with Melodyne (Celemony) is its ability to remove the noise residue that occurs when I record Realivox Blue . . .

Even when Realivox Blue is not singing anything, there is a continuous stream of low-level noise during what should be silent measures, as you can see in the following images of a few measures of the primary melody . . .

Image
Melodyne DNA for Primary Verse ~ Before

Image
Melodyne DNA for Primary Verse ~ After Noise Removed

This is the new version of "Tastes Like Anarchy" . . .

[NOTE: This is a headphone mix; so it's best enjoyed when listening with studio quality headphones like SONY MDR-7506 headpnones (a personal favorite). There now are seven tracks of Realivox Blue singing and plenty of motion-effects and "sparkles" . . . ]

phpBB [video]


Lots of FUN! :)

The Surf Whammys

Sinkhorn's Dilemma: Every paradox has at least one non-trivial solution!
User avatar
by Surf.Whammy on Sun Dec 31, 2017 10:51 pm
I added two tracks of Realivox Blue (Realitone) singing "La-La-La" in the verses, and I increased the input volume levels in the Brickwall Limiter (IK Multimedia) for the two kick drums by +1 dB, but the Brickwall Limiter output levels for two kick drums are not changed . . .

It took a few hours of experimenting to determine the correct notes for the "La-La-La" phrases, mostly because the chord pattern for the verses are more complex than they appear to be, since there's an inversion that changes it from a simple 1-4-5 pattern to something else, which at the moment is the best way I can explain the chord pattern for the verses . . .

One way to explain the chord pattern for the verses is that it's more like "Heat Wave" (Martha & The Vandellas) than "Louie Louie" (The Kingsmen), which makes it what I call an "upside-down Louie Loue", since it's 4-5-1 instead of 1-4-5, where the 4-5 is lower than 1. In 1-4-5, the progression is upward . . .

[NOTE: These two 45 RPM records are monaural, which is the correct way to listen to these songs the way they were heard on AM radio and jukeboxes in the 1960s . . . ]

Image
"Louie Louie" Bass Pattern: 1 = black, 4 = blue, 5 = red

phpBB [video]


Image
"Heat Wave" Chord Pattern

Image
Upside-Down "Louie Louie" Bass Pattern ~ "Heat Wave": 1 = black, 4 = blue, 5 = red

phpBB [video]


[NOTE: This is a stereo headphone mix, and there are plenty of motion effects and "sparkles" that are designed specifically for headphone listening . . . ]

phpBB [video]


THOUGHTS

For the most part, I do everything "by ear"; but I know a good bit of music theory . . .

In this instance, making sense of the song from the perspective of music theory eludes me; so it's a matter of trying different note sequences and then determining what sounds good "by ear", which I do in a ReWire session where Digital Performer 9.51 (MOTU) is the ReWire host controller and NOTION 6.3 is the ReWire slave . . .

The song is coming along nicely, and I am starting to develop an intuitive sense of what to add next, which probably will be some Dubstep deep bass swirls done with Cyclop (Sugar Bytes); some "sparkled" Disco or Trance synthesizer riffs; and perhaps some lead guitar and "sparkled" Latin percussion instruments . . .

I'm also thinking about doing some custom echoes with Melodyne (Celemony) for the tails of various Realivox Blue words . . .

Explained another way, at this point in the development of the song, I am in producer mode . . .

Lots of FUN! :)

P. S. The chord pattern for "Blue Ain't Your Color" (Keith Urban) is similar to the "Heatwave" chord pattern, but it adds a Beatles twist. Instead of the usual {C, Am, F, G} pattern, the Beatles liked to do {G, Am, C, D}, which basically is {4-5-6-1} or {1-4-5-6}, although since the key is G, it's (1-2-4-5}, but so what . . .

The interlude does the drop to "Em-Am" thing that "Heat Wave" does, but instead resolves on C; and in the chorus it does the "Heat Wave" walk-up thing (except in reverse, so it's actually a "walk-down", {C-Bm-Am-G}, but geometrically it's the same pattern according to the way geometry works here in the sound isolation studio), hence it's an upside-down "Louie Louie" with a Beatles twist and a reverse "Heat Wave" walk-up with a modified "Heat Wave" tag, which is fabulous . . .

Surf.Whammy's Pop Song Composing Tip of the Day

"Blue Ain't Your Color" = upside-down "Louie Louie" with a Beatles twist and a reverse "Heat Wave" walk-up with a modified "Heat Wave" tag

[NOTE: This is an example of the mnemonic system I use here in the sound isolation studio to remember stuff like this; and while it might be a tiny bit strange, it works. To make it even easier to remember, I might call it a "Keith Urban"; since "Blue Ain't Your Color" is Keith Urban's signature song. For composing songs, it becomes a matter of constructing a puzzle using puzzle pieces from a lot of other songs, which is considerably easier than trying to have original ideas . . . ]

phpBB [video]


Fabulous! :P

The Surf Whammys

Sinkhorn's Dilemma: Every paradox has at least one non-trivial solution!
User avatar
by Surf.Whammy on Wed Jan 03, 2018 3:39 pm
In this new version, I added Wobble Bass swirls done with Cyclop (Sugar Bytes) at far-left and far-right, along with a new Twin 2 (FabFilter Software Instruments) synthesizer to enhance the Realivox Blue (Realitone) singing during the verses . . .

[NOTE: The knob at top-left is in motion, which is the way the wobbling is done. This is a custom preset for Cyclop named "CONCEPTICON" and was created by Billy Graziadei. I suppose if I studied it for a few months I could discover what it does and how it works; but it sounds good, which is all I need to understand for the "by ear" aspect here in the sound isolation studio . . . ]

Image
Cyclop (Sugar Bytes) ~ Wobble Bass

THOUGHTS

I found the NOTION 3 score for the basic rhythm section and updated the SampleTank 2 (IK Multimedia) VSTi virtual instruments to SampleTank 3 and solved the puzzle of the chord pattern for the verses . . .

[NOTE: These are the bass notes for the set of chords. It's not the detailed chord pattern, but instead is one bass note to represent each chord in the verses. The progression is accurate, although it's not so easy to discern unless you listen to the bass, which helps a little bit, except that the bass is playing "around" the chords, which changes the chords. The "C" chord is played for several measures, and then the chord changes to an "E♭" for while; and the an "F" chord for a while; and back to the "C" chord. Then there's a "G" chord for a while, followed by a "G#" chord; and then it does the tag ("D" followed by "E") and goes back to the "C" chord. The primary chord for the chorus is "C", but it's varied by the notes the other instruments are playing. There's a lot of instrumental counterpoint, which changes everything at various times . . . ]

Image

The other instruments are playing some notes that are not in the chords; and this changes the way the chords sound and makes the Gestalt different . . .

[NOTE: This is mixed for headphone listening, and there are a lot of motion effects and "sparkles" . . . ]

phpBB [video]


Lots of FUN! :)

The Surf Whammys

Sinkhorn's Dilemma: Every paradox has at least one non-trivial solution!
User avatar
by Surf.Whammy on Thu Jan 04, 2018 3:58 pm
There were some pops and clicks in the previous version; and the Cyclop Wobble Bass swirls were too loud; so I did a new mix and added two kick drums . . .

There were too many tracks in Digital Performer, which was part of the problem; so I consolidated some of the tracks by "freezing" them, which renders the audio and replaces the actual tracks with audio clips. This eliminates the computing done by the various effects plug-ins; but if necessary I can "thaw" the tracks and make adjustments to the raw audio, which is a nice feature . . .

Some of the limiters were set to compress rather than to limit, so I changed them to work as limiters rather than as compressors; and I changed them from M/S to L/R mode, which is what they should have been, except that I didn't notice it until I started doing the new mix and revisiting all the instruments and Realivox Blue singing . . .

The basic rhythm section was done five years ago in 2012, and at the time I must not have noticed the M/S mode settings; but now it's corrected and is L/R . . .

[NOTE: This is a headphone mix, so it's best enjoyed when listening with headphones . . . ]

phpBB [video]


Lots of FUN! :)

The Surf Whammys

Sinkhorn's Dilemma: Every paradox has at least one non-trivial solution!
User avatar
by Surf.Whammy on Sat Jan 06, 2018 5:22 am
I did more producing and mixing, this time focusing on balancing the instruments and Realivox Blue singing to focus everything and to make the singing more prominent . . .

THOUGHTS

One of the things I did to focus the instruments was to replace one of the FabFilter Software Instruments Twin 2 synthesizers with a Twin 2 Low Bass synthesizer playing half-notes in a simple bass line with one half-note duration bass note at the start of each measure . . .

The Twin 2 Low Bass pattern is focused and simple; but the IK Multimedia SampleTank 3 Höfner Beatle Bass is there, as well, although it's playing more notes; and there are additional bass synthesizers, which is the producing strategy I use to create deep and rich bass . . .

[NOTE: These are the bass notes for the chords in the verse pattern, shown only as single notes. The chords repeat, but this is not shown. The chorus bass notes all are "C". The notes are shown as "Middle C" in the music notation, but they actually are lower and are bass notes. I had to look at the music notation to discover the simple bass pattern, since it wasn't so obvious due to the Höfner Beatle Bass playing melodic bass lines and the rhythm guitars playing a virtual festival of chords and counterpoint. You would think I should know what the chords are; but that's not the way it works here in the sound isolation studio. I do nearly everything "by ear", and for chord patterns this maps primarily to stacking notes on stems and moving them around until it sound good. In this song, there are three electric guitars playing different sets of chords and counterpoint phrases--and one of them is "sparkled"--which makes it all the more difficult to determine what actually is happening with respect to chords from the perspective of music theory . . . ]

Image

After a good bit of experimenting, I switched some of the limiters to M/S mode for the Realivox Blue singing and adjusted all the singing tracks to focus the singing . . .

The Realivox Blue singing is more focused and prominent in this mix; but I think it's going to be necessary to "duck" some of the instruments to move the Realivox Blue singing in the verses to the front, where the goal is that in the verses, Realivox Blue will sound are realistic as possible . . .

I also need to add a few more over-enunciated ending consonants on a few of the words in the verses, mostly the "stah" ending consonants for a few of the "tah-stah" ("taste") words and a few of the "kuh" ending consonants for "like" so they don't sound like "iie" . . .

One perspective on this is that I'm giving Realivox Blue a Southern accent, which is a personal favorite here in the sound isolation studio . . . :+1

[NOTE: This is a headphone mix, so it's best enjoyed when listening with studio quality headphones . . . ]

phpBB [video]


Lots of FUN! :)

The Surf Whammys

Sinkhorn's Dilemma: Every paradox has at least one non-trivial solution!
User avatar
by Surf.Whammy on Sun Jan 07, 2018 7:01 pm
After listening to some T-ARA songs for a while, I decided to add some digitally synthesized cat purring to Realivox Blue with a FabFilter Software Instruments Twin 2 "Buzz Bass" . . .

[NOTE: Sam Phillips put slapback echo on Elvis. I put digitally synthesized cat purring on Realivox Blue . . . :P ]

Some of the notes in the primary Realivox Blue vocal track for the verses were too nasal; so after doing some experiments, I discovered that by lowering the volume levels of individual DNA blobs, it's possible to smooth the singing level on a specific note-by-note basis, which is even more detailed than using a limiter or compressor. It takes a while to do the DNA blob amplitude edits, but it solves the problem . . .

[NOTE: In Melodyne (Celemony), audio tracks are analyzed using proprietary algorithms that make it possible to identify individual notes and the characteristics of each note. The technology is called "Direct Note Acess (DNA)" and the individual notes are represented visually as "blobs". In the following example, the selected note was too loud; so I used the Amplitude Tool to lower its volume by 2 dB . . . ]

Image
Melodyne ~ Amplitude Tool

I also used several instances of Pro-C (FabFilter Software Instruments) to "duck" the digitally synthesize cat purring (Twin 2 "Buzz Bass"), Hofner Beatle Bass, and Kick Drums based on the activity and volume levels of the primary Realivox Blue singing in the verses and one of the original SampleTank 3 (IK Multimedia) Kick Drums, which was done with SampleTank 2 but now is a Legacy sampled sound library that I imported to SampleTank 3 . . .

I added two Addictive Drums (XLN Audio) Kick Drums, and these are "ducked" using the volume level of one of the SampleTank 3 Kick Drums . . .

The Sample Tank 3 Kick Drum also "ducks" the Hofner Beatle Bass to make room for the Kick Drums when there is a Kick Drum note . . .

All the "ducking" is graceful, so it's subtle but important . . .

THOUGHTS

I revisited the L/R and M/S modes for some of the limiters and "ducking" compressors, and switched a few of them from L/R to M/S so I could control the Mid and Sides volume levels . . .

This is important, because in L/R mode equally balanced Left and Right channels essentially map to positioning the sounds at top-center or middle, while M/R mode lets you control the sides separately from the middle . . .

For the digitally synthesized cat purring, I wanted the sounds on the side; so for the Pro-C (FabFilter Software Instruments) "ducking" compressor, I set it to M/S mode and lowered the volume level for the Mid, but increased the volume level for the Sides . . .

As noted, the primary Realivox Blue vocal track for the verses controls the Pro-C "ducking" compressor for the digitally synthesized cat purring . . .

[NOTE: This is a headphone mix, so it's best enjoyed when listening with studio quality headphones. I "pumped" it a tiny bit with the IK Multimedia T-RackS 5 Brickwall Limiter, so it's hotter than the previous version, but it's not "pumped" so much as most K-pop songs. You can hear the digitally synthesized purring most clearly during the fade at the end of the song . . . ]

phpBB [video]


Lots of FUN! :)

The Surf Whammys

Sinkhorn's Dilemma: Every paradox has at least one non-trivial solution!
User avatar
by Surf.Whammy on Mon Jan 08, 2018 3:18 pm
After listening to the most recent version for a while, I noticed there is an arpeggio soon after the first verse starts. It's subtle, but it's there . . . :)

So, I decided to enhance it with a lead guitar, and I did this for a few more logical places in the verses . . .

[NOTE: I think these lead guitar phrases are called "leading", but perhaps not . . . ]

Image
Subtle Lead Guitar Phrases ~ NOTION 6, SampleTank 3 (IK Multimedia)

Next, I revisited the phonemes for the CYBORG lyrical phrase that occurs after the verses and the "A-N-A-R-C-H-Y" phrases in the chorus, as well as it the end of the song . . .

I experimented with different variations in Melodyne (Celemony) and found one that works nicely, so now it's the CYBORG lyrical phrase throughout the song, which is something you can do in Melodyne, since Melodyne has its own Clipboard and supports cut, copy, and paste, which among other things makes it possible to create custom echoes and is the way I added echoes for the "guh" phoneme at the end of "CYBORG" . . .

[NOTE: I created the one-octave higher voice by copying, pasting, and then dragging upward. Melodyne changes the pitches automagically, which is very nice. The words and phonemes are annotations and are not in the actual audio that Melodyne has analyzed and converted to DNA blobs. In the previous versions of the song, I used a set of pronouns to match the verses {she, you, I, we}, but for the chorus (a) it's too complex and (b) a chorus generally is supposed to be the same throughout the song, so I made all of them "We are CYBORG" . . . ]

Image
"We are CYBORG" Phrase ~ Melodyne (Celemony)

THOUGHTS

I always listen to songs as I am developing them, and after a while I start to hear subtle things, which I then make more obvious . . .

It's a "by ear" strategy, and everything fits together nicely when you listen with the goal being to identify what is there, which always is more than what you recorded . . .

It's a Gestalt . . . :ugeek:

This YouTube video has the Realivox Blue singing soloed, along with a SampleTank 3 (IK Multimedia) Höfner Beatle Bass and a pair of Studio Tight Kick Drums . . .

At present, there are nine tracks of Realivox Blue singing . . .

Some are doubling and tripling; some are harmonizing; and one is singing the "A-N-A-R-C-H-Y" phrases in the choruses . . .

All the Realivox Blue tracks are stereo, and the "A-N-A-R-C-H-Y" track has custom harmony done with Melodyne, as do the "La-La" tracks. The "La-La-La" tracks are "sparkled" and alternate from far-left to far-right . . .

All the Realivox Blue tracks have EQ filters, limiters, Timeless 2 custom echoes; and a few of them are processed with Reel ADT (Waves) to do doubling. The "DJ record scratch" effect is done by one of the Timeless 2 (FabFilter Software Instruments) presets, which was a bit of serendipity . . .

So far, I have worked on the Realivox Blue singing for approximately 75 hours, some of which was focused on discovering how everything works . . .

It takes a while at first, but I think the singing sounds realistic, which is great . . .

Once I discover all the rules, I will be able to do this faster . . .

phpBB [video]


This is the new version with all the instruments and singing . . .

[NOTE: This is a headphone mix, so it's best enjoyed when listening with studio quality headphones . . . ]

phpBB [video]


Lots of FUN! :)

The Surf Whammys

Sinkhorn's Dilemma: Every paradox has at least one non-trivial solution!
User avatar
by Surf.Whammy on Sun Jan 14, 2018 2:05 pm
I did a bit more producing and mixing in this new version, and it's warmer . . . :)

On the producing side, I changed the first few measures of the Intro to make it more dramatic by focusing it on one of the electric guitars and adding two Realivox Blue phrases using phonemes I copied and pasted from the verses in Melodyne (Celemony) . . .

I adjusted the FabFilter Software Instruments Saturn mastering processor to make the mix warmer by lowering the levels of the upper-midrange and high-frequencies, where in this context I am using "upper-midrange" to refer to the "sweet spot" of human hearing in the range of approximately 2K-Hz to 5K-Hz where human hearing is particularly sensitive, which means that frequencies in this range are easier to hear, so they need to be lower in the mix, which is in contrast to deep bass, which is harder to hear and needs to be louder, as explained visually by the standard equal loudness curve (see below) . . .

[NOTE: The "sweet spot" is the midrange-cip in the oval in the equal loudness curve diagram. The way to read the information is that for example, on the threshold curve (lowest curve in the stack), to be perceived as being equally loud, a 20-Hz deep bass note needs to be approximately 70 dB SPL, while a note in the "sweet spot" (a.k.a., "midrange dip") needs to be only -6 dB. For reference, when studio monitors are calibrated to have flat response for the equal loudness curve, notes at the same volume level will be perceived as being at the same loudness, which is the reason for calibrating studio monitors; but at present I think it is better to compensate for the "sweet spot" and generally to lower the ultra-high frequency response. YouTube cuts everything above 15K-Hz, so I also do this. The ultra-high frequencies mostly are annoying and few people can hear them in any immediately conscious way. Most folks imagine they can hear ultra-high frequencies, but for the most part it's a delusion . . . ]

Image

Equal Loudness Curve ~ Wikipedia

Image
"Tastes Like Anarchy" ~ FFT Spectrum Snapshot

Image
"Tastes Like Anarchy" ~ Saturn Mastering Plug-In Snapshot

I redid the recording of the second pair of kick drums, which are done with Addictive Drums (XLN Audio) and changed them to Ludwig kick drums, since the previously recorded tracks (a) were out-of-sync with the pair of SampleTank 3 Studio Tight kick drums and (b) were too deep, since they were different types of kick drums. The Ludwig kick drums are brighter and more focused . . .

THOUGHTS

The last section of the song--where "A-N-A-R-C-H-Y" repeats over and over--needs more stuff, so this is on the producing "To Do" list, as is creating some custom echoes for the Realivox Blue singing, which might include adding backup counterpoint singing . . .

I also am planning to add more "sparkles" and some essentially random "in-and-out" bits based on the observed fact that hit songs tend to have more essentially random bits than most folks realize . . .

For example, in one of the chapters in my new book on digital music production--which is coming along nicely and at present is approximately 425 pages--I devoted a few hours to identifying and counting some of the "in-and-out" bits Michael Jackson added to his signature song "Billie Jean" . . .

[NOTE: These are the times based on the iTunes single version of "Billie Jean" (Michael Jackson), so they are approximately 2 seconds earlier than the YouTube video link (see below). This is not the complete list of "in-and-out" bits, but it's enough to get a sense of what happens in the first 2 minutes and 30 seconds of the song. It does not include the instrumental "in-and-out" bits, which are even more complex and apparently random. The audio on this YouTube video of "Billie Jean" is "pumped", so be careful with the listening level . . . ]

0:27 hiccups
0:37 hiccup
0:42 hiccup
0:49 breath
1:02 hiccup
1:08 hiccups
1:16 hee
1:22 ow ow
1:23 ow ow
1:29 uh
1:34 oh baby
1:39 oooo
1:40 half hiccup
1:42 oh baby
1:46 hee-hee-hee
1:48 hiccup
1:49 hee-hee-hee
1:50 whoo
2:03 hee
2:04 hiccup
2:12 don't think twice
2:13 don't think twice
2:14 oh oooo
2:23 oh no
2:27 hiccup
2:29 oooo
2:29 hee-hee-hee


phpBB [video]


[NOTE: This is a headphone mix, so it's best enjoyed when listening with studio quality headphones. This is not "pumped", but it will be later! :P ]

phpBB [video]


Lots of FUN! :)

P. S. I was watching some TED YouTube videos about music last night, and I discovered that Irving Berlin played everything in the key of F#, which is the one with all the black keys on the piano, as well as a few white keys. Additionally, Irving Berlin did not read or write music notation, so he literally was a "by ear" composer . . .

phpBB [video]


Irving Berlin Insights (Cecil Adams)

Transposing Piano ~ Wikipedia

You can do this in NOTION 6; and among other things it makes it possible for me to do everything with treble clef staves generally in the key of C Major (or one of the related modes that are all white keys on a piano), which is part of my intuitive "by ear" strategy, although I read and write music notation--except that for the most part all that is intuitive is soprano treble clef, since it's what I learned when I was in a liturgical boys choir . . .

[NOTE: I use accidentals (flats and sharps) when it's necessary, but not in the key signature . . . ]

Apparently, I'm not the only one who does this, which is fabulous . . .

Fabulous! :+1

The Surf Whammys

Sinkhorn's Dilemma: Every paradox has at least one non-trivial solution!
User avatar
by Surf.Whammy on Wed Jan 17, 2018 4:55 am
After listening to the now previous version for a while, I decided to add ending "ee" phonemes to the "anarchy" words Realivox Blue sings in the verses . . .

[NOTE: These are two phrases where the first one is the "ee" phoneme that moves downward by a whole step. Instead of "anarchy" ending on "kee", it now ends on "ee", which is more natural. From a practical perspective, I have sent Realivox Blue to The Surf.Whammy Vocal Training School for a while . . . :P ]

Image
"ee" Phonemes Added to "anarchy" ~ Melodyne (Celemony) DNA Blobs

I changed the phrasing of the pair of "taste" words in the first few measures of the Intro; and I adjusted the "ducking" and frequency range of the Höfner Beatle Bass throughout the song, so that the side-chain source for the Mid is increased by 6 dB and now is +12 dB, which lowers the bass in the Mid part of Mid-Side when there is a kick drum note, so that the kick drums have more sonic space. It's a bit subtle, but it works . . .

[NOTE: When the kick drum plays a note, the Höfner Beatle Bass volume temporarily is lowered, which makes sonic space available for the kick drums. There are four kick drums, but only one of them is used for the side-chain source for Pro-C. This happens rapidly, so it's subtle but effective . . . ]

Image
Höfner Beatle Bass ~ "Ducking" via Pro-C (FabFilter Software Instruments)

[NOTE: This frequency range filter restricts the Höfner Beatle Bass to 40-Hz to 3,000-Hz, which primarily cuts the lower and subsonic bass portion and just a tiny bit of the upper-midrange, but all of the high frequencies. This makes sonic space available for the other bass instruments, and it's lower than the lowest note on an electric bass guitar, so nothing that matters is lost . . . ]

Image
Höfner Beatle Bass ~ Frequency Range Filter via Pro-Q (FabFilter Software Instruments)

Some of the "you" pronouns in the 2nd verse sounded more like "we", so I adjusted them to make them sound more like "you" . . .

THOUGHTS

The ending phonemes and notes for singers are very important, and after studying this for a long time, I realized several years ago in an immediately conscious way that there are two things singers do, either (a) due to professional training or (b) naturally:

(1) melodic breathing

(2) embellishing the endings (or "tails") of words

Both of these are related, and sometimes (2) is more a matter of (1) . . .

It's easy to hear when you know to listen for it, and it helps to listen with studio quality headphones for things like this, which can be a bit subtle at times and not so easy to hear when you are listening with calibrated, full-range studio monitors . . .

[NOTE: The various phrases Dion sings often end with an "ah" or an "uh", or some variation thereof, perhaps with a starting consonant like "sah" or "zuh", all of which is (2) . . . ]

phpBB [video]


[NOTE: Elvis Presley was very skilled in doing this, and if you listen carefully you can hear him hiccuping in a few of his melodic breaths. He did this naturally, but I think he also had some advice on melodic breathing from his various producers or perhaps some of the professionally trained singers who worked with him, like The Jordanaires. who would have learned this as part of Gospel singing, which also was a favorite style for Elvis. This also is a huge clue that Elvis was not singing loudly. Instead he was singing pianissimo or even lower volume at times, which is one of the ways he "worked" a condenser microphone and the limiter and plate reverb. When he needed to sing fortissimo, he backed-off the condenser microphone considerably, which is another skill singers learn in the recording studio. You know he is singing very softly, because his breathing is nearly as loud as his singing at times. It sounds loud, but that is a producing and audio engineering technique which is best done optimally when the singer knows how the recording and producing techniques work . . . ]

phpBB [video]


[NOTE: The Everly Brothers are personal favorites, and they do (2) very obviously . . . ]

phpBB [video]


This is the new version of "Tastes Like Anarchy" . . .

[NOTE: It's a headphone mix, so it's best enjoyed with listening with headphones. Later I will switch to mixing when listening to playback through the calibrated, full-range studio monitor system, but for what I am doing now, headphones work best. This is not "pumped" . . . ]

phpBB [video]


I have discovered that Melodyne has its own Clipboard for each instance, which means that you cannot copy DNA blobs from one instance of Melodyne and then paste them to another instance of Melodyne . . .

The practical consequence is that you need to add some music notation and phonemes to the singing, typically beginning several measures after the actual end of the song, so that later you have specific consonants and vowels to copy and paste to enhance or "humanize" the Realivox Blue singing . . .

Fortunately, I did this to some extent without thinking about it . . .

The other way is to record a new set of phonemes on a different staff and track; but that's more complicated than having the phonemes available in the track you need to enhance or "humanize" . . .

Lots of FUN! :)

The Surf Whammys

Sinkhorn's Dilemma: Every paradox has at least one non-trivial solution!

17 posts
Page 1 of 1

Who is online

Users browsing this forum: No registered users and 5 guests