Afrikaans
Akan
Albanian
Amharic
Arabic
Armenian
Azerbaijani
Basque
Belarusian
Bemba
Bengali
Bihari
Bosnian
Breton
Bulgarian
Cambodian
Catalan
Cebuano
Cherokee
Chichewa
Chinese (Simplified)
Chinese (Traditional)
Corsican
Croatian
Czech
Danish
Dutch
English
Esperanto
Estonian
Ewe
Faroese
Filipino
Finnish
French
Frisian
Ga
Galician
Georgian
German
Greek
Guarani
Gujarati
Haitian Creole
Hausa
Hawaiian
Hebrew
Hindi
Hmong
Hungarian
Icelandic
Igbo
Indonesian
Interlingua
Irish
Japanese
Javanese
Kannada
Kazakh
Kinyarwanda
Kirundi
Kongo
Korean
Krio (Sierra Leone)
Kurdish
Kurdish (Soranî)
Kyrgyz
Laothian
Latin
Latvian
Lingala
Lithuanian
Lozi
Luganda
Luo
Luxembourgish
Macedonian
Malagasy
Malay
Malayalam
Maltese
Maori
Marathi
Mauritian Creole
Moldavian
Mongolian
Myanmar (Burmese)
Montenegrin
Nepali
Nigerian Pidgin
Northern Sotho
Norwegian
Norwegian (Nynorsk)
Occitan
Oriya
Oromo
Pashto
Persian
Polish
Portuguese (Brazil)
Portuguese (Portugal)
Punjabi
Quechua
Romanian
Romansh
Runyakitara
Russian
Samoan
Scots Gaelic
Serbian
Serbo-Croatian
Sesotho
Setswana
Seychellois Creole
Shona
Sindhi
Sinhalese
Slovak
Slovenian
Somali
Spanish
Spanish (Latin American)
Sundanese
Swahili
Swedish
Tajik
Tamil
Tatar
Telugu
Thai
Tigrinya
Tonga
Tshiluba
Tumbuka
Turkish
Turkmen
Twi
Uighur
Ukrainian
Urdu
Uzbek
Vietnamese
Welsh
Wolof
Xhosa
Yiddish
Yoruba
Zulu
[Music]
I like to start with a very basic
backing track to create the midi for the
vocal Melody
[Music]
I could export this as a midi file but
I'm actually going to drag and drop it
straight into synthesizer V
but before I do
just a little tip here
as much as the playing this in
felt right it's actually quite good to
have all your synthesizer V notes as
bolted up to each other and Legato
unless it really needs a specific Gap
and of course you can always adjust that
once you're in there but just to save
time I like to just fill all those in
okay inside synthesizer V the very first
thing you should do which is something I
always forget to do because it's just
that little number there
is to set the tempo
of the track you're working on and I
know it's 90. and then the other thing
which is weird is I want to import my
backing track but it only allows you to
import midi tracks here what you have to
do is right click here and go create a
new instrumental track
and then I can load in
audio track let's see
[Music]
now let's go back to Cubase and get the
solarius MIDI
and I can actually just drag that
cross let go in here and I want to make
a new track
there we go and it's not the one with
zero notes so yes okay
[Music]
so if you're just getting a sine wave
tone it means there's no voice loaded
I'm going to delete this track here and
I need to assign Solaria to there
and then I surrendered them and that was
quick
[Music]
okay so as you see there there was notes
missing sound wise that means we have a
conflict because obviously a vocal is a
monophonic source so there must be an
overlap here
make that a bit wider
so let's just check
oh that's this one
and that one
[Music]
la la la
[Music]
la
la la la la la
okay so let's get some words in there
you can select the whole song or you can
just select one line at a time so start
with this
and you go to insert lyrics
and instead of la la as we say
okay so you can already see a problem
believe should be over two pieces not
over one
what you can do is you can
type believe in here
and then using the Tab Key you can just
go
plus go to the next word
and then put you
and then
set okay try again
I can't believe you said
[Music]
okay not bad so having the words butt up
against each other like this
is works the best leaving gaps even
if you feel like you need a gap there
more often than not singing you you
don't really leave them unless it's a
very clear Gap
and then you can always insert that all
you need to do is just drag it smaller
and it creates a gap so
you can always edit all of that once
you're in there so this best starting
point is to start like this
and then I'm going to add add the next
lines in here
but this time what I'm going to do is
actually
because they go
[Music]
okay I put all ways separate like that
just to make it easier for this to
figure out where they're going
I have them all in here if I go to the
next one and go Plus
plus and plus
plus plus
it spreads them out and you can see the
phonemes here how it's actually using to
make the pronunciations
[Music]
let's just mute this for a sec
we can always be good friends
and the reason that seems to work better
is because
there are certain words that will always
always be it's almost like there's no
separation between the Z and the B
so if you have it as a continuous string
of words the algorithm just does a
little more natural version of it so the
next one I'm going to do word by word
again just so we can hear the difference
your words spin around my head
just change this
your words spin around my head if we
were to go
your
words
and then put a Plus on the next one
your word spin round my head
your words spin round my head
I'm gonna now put the last phrase in all
in one go again
remember
so putting pluses in everyone
see what this does like a song that
never
supposed to hold that note Why didn't it
well because this is not a new
syllable
so I put a minus instead
and that extends that so she now you see
that this holds this note to the end of
there like a song that never
[Music]
and if you want to say split a track and
do some kooky like um
Boca flourish type of thing
[Music]
when using minus instead of the plus
allows you to use a sustain note and
change the notes within it
like a song
I'm hearing something kind of weird like
a pitch sweep on
that on that
song that never
and you can see these massive
pitch swings so you can actually go in
and redraw these lines
I'm going to smooth that out a bit more
and that one doesn't want to go down to
kill back up again so let's just do drag
it like that
and this should probably be rising up
okay they're a little bit wobbly at the
moment but you'll you'll see what we can
do about that
like a song that never
so down below
there are a bunch of graphs
pitch vibrato loudness tension
breathiness voicing gender tone shift
pitch deviation shows us the edits we
just made
and you can
basically draw them here and see them
moving up there
and like this one here has got a lot
going on
and what's actually happening is
the vibrato that you're seeing like here
that's
the singer's natural vibrato
it's not part of the synthesis
so in order to fix that you have to
counteract it
so if I want to reduce this vibrato a
little bit the easiest way to do it is
to draw it first over top just
not as big
that gives you crazy control and being
able to you know go slightly sharper
flat
at the beginnings of words or when you
know when the singer's belting it out
they always tend to go a little sharp
first
whereas this has gone nice and smoothly
there so I could actually draw that up
and
and then bring it back like that see how
that sounds
[Music]
yeah not bad
and like I said you can go in here and
edit your heart's content
it's just affecting what you're seeing
on the top line there and in some ways
it's easier like for instance if I just
wanted to make this whole section rise
I could use a line tool and just grab
that draw it say up like this
and it's gonna be bad like a song that
never
get the idea so the other things that
are down in the edit here the vibrato
envelope again this is back from when it
was just the synth voice not the AI the
AI voices don't respond to all of these
things loudness they definitely do
tension they do so on this particular
note if I wanted to you know make the
tension get stronger as as it builds
I can draw a line in from start of that
and actually I'm going to push it right
to the top see how it's where that gets
us
it's not going to be a huge thing it's
just going to be subtly more like I said
[Music]
and to illustrate that the easiest thing
to do is to
do it the opposite way
you'll hear the voice
reduce
to like a breathy voice
[Music]
which is really powerful once you get
going on this you can
have a lot more control than you imagine
from from the get-go
so let's put the music on and just have
a listen to what we've got so far
[Music]
like a Song That Never
Ends okay are are weird that is still
messing around
so let us get rid of this Segway
or reduce the the dependence on it
by going back in here
and clearing these words out
and then starting fresh here
okay so remember tab next one I want a
plus next one I want to plus
next one I want to plus
and then the rest will want minuses
okay I see what this does
[Music]
[Music]
[Music]
something else that is good to know is
let's put the snap back on
and with a pen tool like her or the
pencil tool I can draw in a new word
and in this time I'm going to draw in a
la
but what I want there is a breath so you
put br
and then
take the music off
like a song that never
[Music]
and you don't have to do it all the time
but sometimes it just helps the first
word
it tends to flow a little better a
little more natural
let's review
[Music]
that that is getting on my nerves
so I'm going to get rid of the other
words in there see if that helps
[Music]
I don't know why it's glitching on the
ends
if you like you can try this
I just discovered this
you can
similar to the breath
just put in HH
have a listen to that
it's it's like a breath but it seems not
as aggressive as the first one so let's
stick that on the end here
all right let's see what this sounds
like
[Music]
just now I realize that that is way too
short
so here we go back to our phonemes
what we want to extend is this that part
because it's got half the word is the to
let's go in here and extend that as much
as we can
so let's go minus so let's see if that's
any better
like a song
ever
definitely better we have a strange huge
pitch dip there
so this is where we draw in
I'm going to take it from here
like a song that never
I'm also gonna tune this up
like a song
someone gets a bit sharp at the end
like a song that never
that sounded good
I can't believe you said all right let's
turn the music off
so if you go back to the voice you'll
see that this is her default voice which
is clear
I can't believe you said
and if you go along
next one is soft I can't believe you
said
and then Airy
I can't believe you said
pretty similar to
I can't believe you said
it's power
passionate
I can't oops let's give it a chance to
render I can't believe you said
they sound similar but there's
definitely differences between clear
soft and Airy sounds similar power and
passion it sounds similar what's solid
like I can't believe you said
feels a little different and then light
I don't fully understand
I can't believe you said okay so it's
toned down a little bit
you have all these great singing Styles
which again increases the power hugely
and the fun begins when you can actually
mix you can mix them up you can design
your own sound so you could blend in
kind of passionate 100 and 50 Airy
something like
I can't believe you said
okay which is different than clear
I can't believe you said you really hear
it on the belief on the Eve
I can't believe you said
all right so let's go about what do we
do
50 percent
like that I can't believe you said
which is awesome
and then it gets better
and that would be nice if you get real
time animate these but you can't that's
what these other parameters are for
but there is also a way to vary the
performance and that's what takes and
you have pitch takes and tomber takes
and I can select the whole
phrase or the whole vocal if you like
and just generate a new take calls it
take one
and now I can control between the
default and more expressive version
so if you just watch because this is
pitch watch the wiggly lines
if I wanted to just change this last
note and I can generate a Timbre
take actually I'll do them all I'll take
them here generate a take and I put this
down to default
I can't believe you said
if you said
you said
so yeah there's a little more power
behind that one
what if I go the other way
said
so it's softer which actually sounds
like tick the default tick
said no it's somewhere in between
clue is when you watch the waveforms
when they redraw off their different
shapes I can't believe you said
so that's quite nice
so there's just one
at the moment I can just say that tomber
that's generated take
okay so now they're all take twos
oh yeah
let's go the other way this one was this
way I think
I can't believe you said
I can't believe you said
awesome
and then just a final explanation on
these graphs down here the one on the
right is just a display the idea being
that if I've done a pitch deviation like
this
and I want to do a loudness to go with
it I can see it there I can basically
well this happens to only be at the end
of the note but let's get rid of that
anyway so you can view one thing over
here well let's do let's do the tension
okay and if I just use the line tool
here
and I'll make it get a little more tense
that I can't believe you said and now
I'll leave that on tension and then this
one I'll switch to loudness and see I've
got that there now so if I want them to
rise together or very similar I can just
use that as a guide
I can't believe you said there you go I
think this is an awesome tool it's early
days still and it's already this good
imagine what it's going to be like in a
year from now
[Music]
Can't find what you're looking for?
Get subtitles in any language from opensubtitles.com, and translate them here.