marty
🏍️ Giga poster
Human computer voice? Less annoying that a real human female.
Can these talking computers ya want to buy convert text to voice? Like this ↓
-
Hello ES! We could use some help to get us past the finish line on building the new knowledgebase for the forum.
Can you donate? Please see our fundraising page. Thank you!
Amberwolf's Music Studio Technical Stuff (and the sonic products thereof)
- Thread starter amberwolf
- Start date
Can these talking computers ya want to buy convert text to voice? Like this ↓
Turns out a fair number of the types of sounds (vocals) I was after were already available free within various free bundles posted over the years on Black Octopus and Ghosthack. Ethnic Voices has a lot of the vocal style I wanted out of Vocalisa and doesn't require buying Kontakt for $300 on top of it, though it does have a Kontakt instrument included, it also has the sounds available separately as wave files I can just use directly. That makes it more complicated to create the vocals but at least it's really really cheap compared to the other way. 
It's not as sophisticated a set of sounds, but it will let me experiment, and do at least some of the things I wanted to try.
So I did go ahead and buy some of the sets, and built this song out of some of the stuff in some of them, plus other samples I already had (or have made previously myself), plus the Z3TA2+ synth (so far for just the bassline, but it'll get used for a bunch of other stuff before I'm done).
Current version:
https://www.soundclick.com/amberwolf/?more=14881894
Original version:
https://www.soundclick.com/amberwolf/?more=14881842
For those willing to listen, I'm looking for detailed / helpful critiques of any part of the track, technical, musical, or otherwise. Give it to me straight.
Stuff I already know about that I'm working on: There's too much going on in several places, and some of the transitions are pretty abrupt. It has a LOT of work still to go, but I have a direction to go in, at least. I expect it will end up at least twice as long as it is now with the interludes and intro stuff I haven't built and played yet. Many of the samples in the libraries need trimming / chopping up / key changing in various ways to create the melodies and harmonies I'm after (I already did some of this for certain sections, but there's a lot yet to go). The percussion I still have to build transitions and "thinner" sections of to clear out some sections where there's too much going on.
But any feedback about those is also still welcome.
I wish the libraries had a "lyrics" translation for the non-English vocals. I love the sound of them but it would be nice to know what they mean.
(one of them has translations for the Elvish parts, but nothing for the real languages).
It's not as sophisticated a set of sounds, but it will let me experiment, and do at least some of the things I wanted to try.So I did go ahead and buy some of the sets, and built this song out of some of the stuff in some of them, plus other samples I already had (or have made previously myself), plus the Z3TA2+ synth (so far for just the bassline, but it'll get used for a bunch of other stuff before I'm done).
Current version:
https://www.soundclick.com/amberwolf/?more=14881894
Original version:
https://www.soundclick.com/amberwolf/?more=14881842
For those willing to listen, I'm looking for detailed / helpful critiques of any part of the track, technical, musical, or otherwise. Give it to me straight.
Stuff I already know about that I'm working on: There's too much going on in several places, and some of the transitions are pretty abrupt. It has a LOT of work still to go, but I have a direction to go in, at least. I expect it will end up at least twice as long as it is now with the interludes and intro stuff I haven't built and played yet. Many of the samples in the libraries need trimming / chopping up / key changing in various ways to create the melodies and harmonies I'm after (I already did some of this for certain sections, but there's a lot yet to go). The percussion I still have to build transitions and "thinner" sections of to clear out some sections where there's too much going on.
But any feedback about those is also still welcome.
I wish the libraries had a "lyrics" translation for the non-English vocals. I love the sound of them but it would be nice to know what they mean.
here's version 3 with some more accent vocals, percussion, and string bits added, and some main percussion altered or removed, some transitions in various tracks shifted to better line up with each ohter.
110124 000001 000038n
110124 000001 000038n by Amberwolf
110124 000001 000038n
110124 000001 000038n by Amberwolf
Are you accepting critiques, or have you finished your education for the time being? Infinite improvement would translate to either being me lending a helping hand.
Then again, credit networks may not be capable of evaluating the veracity of my identity. Strike all of that, never mind!
Addendum: grade received was C for outcome, A for effort, A for expected longevity. Sympathy demanded a D. No other records available. "Want a footnote like this on your content, too? Inquire within!"
7th new version;
110124 000001 100092G
110124 000001 100092G by Amberwolf
I experimented and built a bass-companion track out of tiny slices of a deep female voice from the samples out of the free Marie-Pneuma instrument from PianoBook, to complement the synthbass (Z3TA2+) bassline in some places. It was going to replace that track, but didn't do what I wanted everywhere, so it's just addon in places and replacement in a very few spots. Rather than use the clip's pitch dialog, I used Kerovee (automatable tuner/pitchshifter)as a track FX and automation to control the transpose to match the synthbass notes, after first building "clips" of the arp pattern in the voice by using the clip pitch dialog then bouncing the resulting many tiny clips into 2-measure-long sections to use and slice up as needed.
Also further edited the vocal tracks to try to cure the repeititititiveness; that's still a WIP though it's much better. More mix changes, etc.
110124 000001 100092G
110124 000001 100092G by Amberwolf
I experimented and built a bass-companion track out of tiny slices of a deep female voice from the samples out of the free Marie-Pneuma instrument from PianoBook, to complement the synthbass (Z3TA2+) bassline in some places. It was going to replace that track, but didn't do what I wanted everywhere, so it's just addon in places and replacement in a very few spots. Rather than use the clip's pitch dialog, I used Kerovee (automatable tuner/pitchshifter)as a track FX and automation to control the transpose to match the synthbass notes, after first building "clips" of the arp pattern in the voice by using the clip pitch dialog then bouncing the resulting many tiny clips into 2-measure-long sections to use and slice up as needed.
Also further edited the vocal tracks to try to cure the repeititititiveness; that's still a WIP though it's much better. More mix changes, etc.
8th new version,
110124 000001 100103L vocals
https://www.soundclick.com/amberwolf/?more=14896411
eqd out some of the lower tones in the vocals based on the very little feedback so far (all on the cakewalk forum) that said it was muddy, been trying to find other vocals that still fit in the places where the present leads are repetitive, so there are a few changes in that stuff too.
some of the vocal samples I have available to use are not the right notes so i have had to pitchshift parts of them to go where i want. this sounds very artificial. i have been trying to figure out how to use kerovee to tune them more naturally, but it's not intuitive to me.
i don't have the option of any other one since i can't install new things for it (long story) and the "built-in" correction (celemony melodyne? vvocal? can't remember which) is very very hard to use, poorly designed ui, and crashes unpredictably even when you're not doing anything with it (just having it inserted in the project anywhere can be enough). Sometimes you can't even reopen the project file after inserting it even if ti doesn't crash (one type of problem that is a reason I never use save, only save-as, so i always have versions of the project to go back to with only a short time's worth of editing/creation lost).
110124 000001 100103L vocals
https://www.soundclick.com/amberwolf/?more=14896411
eqd out some of the lower tones in the vocals based on the very little feedback so far (all on the cakewalk forum) that said it was muddy, been trying to find other vocals that still fit in the places where the present leads are repetitive, so there are a few changes in that stuff too.
some of the vocal samples I have available to use are not the right notes so i have had to pitchshift parts of them to go where i want. this sounds very artificial. i have been trying to figure out how to use kerovee to tune them more naturally, but it's not intuitive to me.
i don't have the option of any other one since i can't install new things for it (long story) and the "built-in" correction (celemony melodyne? vvocal? can't remember which) is very very hard to use, poorly designed ui, and crashes unpredictably even when you're not doing anything with it (just having it inserted in the project anywhere can be enough). Sometimes you can't even reopen the project file after inserting it even if ti doesn't crash (one type of problem that is a reason I never use save, only save-as, so i always have versions of the project to go back to with only a short time's worth of editing/creation lost).
Guerney
⚡ Regular
- Joined
- Dec 28, 2021
- Messages
- 140
I might have suggested this before: clone the drive, then try a registry cleaner on the clone? If it appears to improve the situation (might of course do the opposite) using a convenient automatic setting, try it for awhile to make sure the result is stable... and if not, try another registry "Cleaner"? You'd always have the old system to return to, particularly if you cloned it onto two backup drives, and if you kept those two drives in two locations a long way apart in case of some unforseen catastrophe, rather than just the one. Functioning secondhand hard drives are two a penny these days - last week I dragged a 500GB out of a neighbour's Sky TV satellite box he was throwing away.
I keep my main music making computer hardware copied with OS and program drive cloned, plus clones as backups, and cloned backups of the data drives too... and I'm about to make a complete copy of a new more modern one, with all it's more current hardware, programs and plugins. This way I can use various VST instruments with control surfaces separately, recording to any of the others, with all sorts of modifications of the noises in between. Hey everyone has gotta have a hobby. Who cares if nobody else gets it.
Analytics page of an old tune re-uploaded 216 days ago:
BTW, mysteriously, Youtube's algorithm is still pushing this tune. Why? I really have no no idea, but very much wish I knew what prompts it's algo to push noises around like this at poor innocent unsuspecting and wholly unprepared Youtube users. The good news is, at this rate there's zero danger of my ever becoming famous, thank goodness. That's the last thing I need right now. It'd be nice if Youtube paid me some share of the advertising they are profiting from these views. This many views might be worth as much as a whole unchewed £1 coin.
Must say, your tracks are sounding good, but there's too much under 20Hz taking up sonic space in this track. Not needed, the human ear can't hear it, most speakers can't reproduce it - I aggressively remove anything under 20Hz without mercy, or as much as possible. The spectra of this track is wrong for playback on the vast majority of speakers, listen with several and you'll hear what I mean, especially cheapo speakers most people use. Also listen with several different headphones. You're shooting yourself in the foot with all this stuff under 20Hz. Look at it, LOOK AT IT!!! It's monstrous! Why Amberwolf? Why?
Last edited:
i'll have to see what's there; i'm not sure my stuff shows me that low but i'll look. i'm sure i have an eq that can roll off that low even if i can't see or h ear it.
10th new version
i doubled the volume of the vocal tracks, brought the strings up with them but only about half that much, and did some more percussion / etc tweaks (couldn't sleep, as usual, but at least the party people are done shaking the neighborhood finally)
110124 000001 100115P
110124 000001 100115P by Amberwolf
10th new version
i doubled the volume of the vocal tracks, brought the strings up with them but only about half that much, and did some more percussion / etc tweaks (couldn't sleep, as usual, but at least the party people are done shaking the neighborhood finally)
110124 000001 100115P
110124 000001 100115P by Amberwolf
so i looked at the master bus wiht voxengo span, witht eh lowest low setting it has of 5hz, and it shows there is some stuff but the spectrum doesn't look anything like what your image shows, it shows it rolling off down there not flat / going up?
either way i went ahead and stuck a 6-node eq in there and set all the nodes for a high q 24 and -18db for a steep notch to get as surgical as i could, and stacked the bands from the zero end to 20hz, so it should be completely taking out the whole area from just above 20hz on down without taking much above that. span doesn't show any difference with it in the masterbus (span is the last thing in the entire chain and bus path), and even if i insert the same eq setup into every bus, it doesn't change the span readout or the waveforms on the buses which are realtime drawn by the playback.
i can't hear a difference either but i'm sure the vizio soundbar i'm listening thru isn't capable of that low a frequency anyway.
i know hte eqs are working because if i take their master gain down it drops the whole output down.
have to use the steep notching stacked up because the lowpass fliter or the shelving modes have a nonchangeable very low q and they roll off lots of stuff way too much above the center frequency, up to 50hz or more above the point it is flitering, and that's with the q at 2. if i up the q it doenst narrow the filter, it just greatly *increases* the output in the octave or more above the filter frequency, so i cant' really use the lp or shelf modes for this; they take too much out and/or change the whole balance too much even using just one band.
i'll have to load up the exported wave file from the master bus into audacity where i can access other newer stuff like filtron and see what that will do (though it's already past the multiband compressor at that point)
either way i went ahead and stuck a 6-node eq in there and set all the nodes for a high q 24 and -18db for a steep notch to get as surgical as i could, and stacked the bands from the zero end to 20hz, so it should be completely taking out the whole area from just above 20hz on down without taking much above that. span doesn't show any difference with it in the masterbus (span is the last thing in the entire chain and bus path), and even if i insert the same eq setup into every bus, it doesn't change the span readout or the waveforms on the buses which are realtime drawn by the playback.
i can't hear a difference either but i'm sure the vizio soundbar i'm listening thru isn't capable of that low a frequency anyway.
i know hte eqs are working because if i take their master gain down it drops the whole output down.
have to use the steep notching stacked up because the lowpass fliter or the shelving modes have a nonchangeable very low q and they roll off lots of stuff way too much above the center frequency, up to 50hz or more above the point it is flitering, and that's with the q at 2. if i up the q it doenst narrow the filter, it just greatly *increases* the output in the octave or more above the filter frequency, so i cant' really use the lp or shelf modes for this; they take too much out and/or change the whole balance too much even using just one band.
i'll have to load up the exported wave file from the master bus into audacity where i can access other newer stuff like filtron and see what that will do (though it's already past the multiband compressor at that point)
Guerney
⚡ Regular
- Joined
- Dec 28, 2021
- Messages
- 140
That dude has better listening equipment than you have, but you can remove unneeded low frequencies to polish the track mostly visually. It's persistent, that muddy woolly<20kHz guff. It builds up with some effects more than others, particularly reverb or synths (my Korg MS20 VST is a low end monster). One or two passes of software EQ doesn't do it, I have to use about 8 or more in an effects chain, at the end of all of the other effects. I do that for every individual track of a multitrack, then again over the whole lot. However many instances of EQ are required to get rid of the horrible stuff, or enough of it. Stamp it down. Then the all important mid range can breathe with more clarity, making those dudes on cakewalk forums shut the hell up.
Last edited:
FWIW, I stacked so many copies of the eq in series on each bus and the master and saw no difference at the low end that something must be wrong with the way i am setting up the eq, but i can't see anyhthing different from how i normally use it except the frequencies I'm cutting out. so...i pulled away from this one for a bit and started:
A new track, A Peek Over The Wall:
This is related to a (very good) anime called Haibane Renmei; imagine one of the Haibane sneaking up Guri's wall, and getting a look outside....
This one I started because I was having trouble hearing anything i needed to in the other WIP above so i forced myself to (so far) just use sounds out of the "beta" free packs from Cymatics.fm, and to do what I could to misuse them by layering any loops so they are from different "sets", if there were "stems" I used them from different sets too, with a few exceptions, to make me make decisions I probably would not otherwise have done. I also "splattered the canvas" with various sounds first, dropping things into the tracks (using a template made from the other WIP).
Except the bassline, which is created using some of the midi clips supplied with some of the stemmed clip sets, then edited / cleaned up to play a Z3TA2+ synth with an arp.
EDIT:
New version 2, 120524 000001 000030h -- mix edits, arrangment tweaks, etc
A Peek Over The Wall, by Amberwolf
original version still over here
A Peek Over The Wall - WIP by Amberwolf
A new track, A Peek Over The Wall:
This is related to a (very good) anime called Haibane Renmei; imagine one of the Haibane sneaking up Guri's wall, and getting a look outside....
This one I started because I was having trouble hearing anything i needed to in the other WIP above so i forced myself to (so far) just use sounds out of the "beta" free packs from Cymatics.fm, and to do what I could to misuse them by layering any loops so they are from different "sets", if there were "stems" I used them from different sets too, with a few exceptions, to make me make decisions I probably would not otherwise have done. I also "splattered the canvas" with various sounds first, dropping things into the tracks (using a template made from the other WIP).
Except the bassline, which is created using some of the midi clips supplied with some of the stemmed clip sets, then edited / cleaned up to play a Z3TA2+ synth with an arp.
EDIT:
New version 2, 120524 000001 000030h -- mix edits, arrangment tweaks, etc
A Peek Over The Wall, by Amberwolf
original version still over here
A Peek Over The Wall - WIP by Amberwolf
A Peek Over The Wall
New version 5
120524 000001 000046l
https://amberwolf.bandcamp.com/track/a-peek-over-the-wall
previous versions still up here
https://www.soundclick.com/amberwolf/?more=14900124
https://www.soundclick.com/amberwolf/?more=14901375
https://www.soundclick.com/amberwolf/?more=14901398
https://www.soundclick.com/amberwolf/?more=14901402
https://www.soundclick.com/amberwolf/?more=14901818
https://www.soundclick.com/amberwolf/?more=14901819
Main changes: removed wierd bellchime intro measures to go straight into the piano, reduced bass a lot, reduced percussion some, effected some percussion clips (reverb to push back in stage).
I also added two string tracks, one Dimension with Full Tremolo strings, the other SI Strings Detache Strings, with hand drawn (snap off for "feel and movement") versions of some of the vocal melodies as backing accents in various places. Synth track volume automation was used for dynamics to "draw up" some notes after they start to go along with the vocal dynamics, or in some cases to contrast with them.
New version 5
120524 000001 000046l
https://amberwolf.bandcamp.com/track/a-peek-over-the-wall
previous versions still up here
https://www.soundclick.com/amberwolf/?more=14900124
https://www.soundclick.com/amberwolf/?more=14901375
https://www.soundclick.com/amberwolf/?more=14901398
https://www.soundclick.com/amberwolf/?more=14901402
https://www.soundclick.com/amberwolf/?more=14901818
https://www.soundclick.com/amberwolf/?more=14901819
Main changes: removed wierd bellchime intro measures to go straight into the piano, reduced bass a lot, reduced percussion some, effected some percussion clips (reverb to push back in stage).
I also added two string tracks, one Dimension with Full Tremolo strings, the other SI Strings Detache Strings, with hand drawn (snap off for "feel and movement") versions of some of the vocal melodies as backing accents in various places. Synth track volume automation was used for dynamics to "draw up" some notes after they start to go along with the vocal dynamics, or in some cases to contrast with them.
Last edited:
Just for the sake of trying it, I rearranged some of the clips and tracks so I could just mute out all the bassline and most of the percussion, leaving just the slow stuff and accents, to make a "closer to ambient" version :
https://www.soundclick.com/amberwolf/?more=14901552
To really become ambient it would need quite a bit of changes, and an underlying unifying bed-sound I haven't decided on. It would need to not only fit what's there, but also fit the idea behind the song. In that show, the opening is full of rushing air / wind as someone falls and falls and falls in their cocoon dream, so, probably that kind of thing, so there's a temporary bed of wind sound.
I'm still working on the original "ambient beats"
version but the above will probably be the basis of an extended full-on ambient version of it.
:https://www.soundclick.com/amberwolf/?more=14901552
To really become ambient it would need quite a bit of changes, and an underlying unifying bed-sound I haven't decided on. It would need to not only fit what's there, but also fit the idea behind the song. In that show, the opening is full of rushing air / wind as someone falls and falls and falls in their cocoon dream, so, probably that kind of thing, so there's a temporary bed of wind sound.
I'm still working on the original "ambient beats"
version but the above will probably be the basis of an extended full-on ambient version of it.
New version 7
120524 000001 000057p
Cleaned out unnecessary bassnotes that trigger the arp, and transposed the remaining ones in various places to match vocals or complement them. Added wind fx (same as that used in "ambient" version) at beginning and just barely in the background during the break. Other minor edits and timing tweaks, various mix tweaks.
120524 000001 000057p
Cleaned out unnecessary bassnotes that trigger the arp, and transposed the remaining ones in various places to match vocals or complement them. Added wind fx (same as that used in "ambient" version) at beginning and just barely in the background during the break. Other minor edits and timing tweaks, various mix tweaks.
If I Should Wake
122224 000001 000021g -- first public version.
At only 30 seconds, it's likely this is just going to be the intro for a long evolving track. This one is another experiment with "misusing" sounds from assorted sources (mostly Ghosthack and Cymatics for this one) and building something around them after dropping them randomly in the track view, then rearranging (a lot) to make it musical, and begin forming what I want out of it, as I begin to hear something coming together.
122224 000001 000021g -- first public version.
At only 30 seconds, it's likely this is just going to be the intro for a long evolving track. This one is another experiment with "misusing" sounds from assorted sources (mostly Ghosthack and Cymatics for this one) and building something around them after dropping them randomly in the track view, then rearranging (a lot) to make it musical, and begin forming what I want out of it, as I begin to hear something coming together.
12-29-24: 122224 000001 100035j -- first extension and transition, using some of the same pieces as the first section (re-edited for the new one) plus some edited (retuned, snipped, processed, etc) parts of clips from Ghosthack's beautiful Shymer collection.
Note the second section is stylistically different from the first; connected by the sounds and some of the musical phrasings. Only adds another minute for a total of about 1 and a half minutes total, enough to leave you wishing for more.
EDIT: 12-30-24: 122224 000001 100042k -- extensive small edits, added sounds, altered pitches, etc.
EDIT: 12-30-24: 122224 000001 200063p -- second extension and transition (this section is more "cinematic"). Still under 3 minutes so far; am working on section four now, which will probably add another minute or so.
If nothing else, it is very pretty.
EDIT: 12-30-24: 122224 000001 200069r -- extensive small edits in all sections
12-31-24: 122224 000001 200071s -- extensive small percussion and timing edits.
EDIT: 1-1-25: 122224 000001 200076u -- extensive small percussion and timing edits.
Note the second section is stylistically different from the first; connected by the sounds and some of the musical phrasings. Only adds another minute for a total of about 1 and a half minutes total, enough to leave you wishing for more.
EDIT: 12-30-24: 122224 000001 100042k -- extensive small edits, added sounds, altered pitches, etc.
EDIT: 12-30-24: 122224 000001 200063p -- second extension and transition (this section is more "cinematic"). Still under 3 minutes so far; am working on section four now, which will probably add another minute or so.
If nothing else, it is very pretty.
EDIT: 12-30-24: 122224 000001 200069r -- extensive small edits in all sections
12-31-24: 122224 000001 200071s -- extensive small percussion and timing edits.
EDIT: 1-1-25: 122224 000001 200076u -- extensive small percussion and timing edits.
Last edited:
1-4-25: 122224 000001 200080w -- extensive small percussion and timing edits, mostly removing percussion tha was "in the way" of other things.
If I Should Wake, by Amberwolf
If I Should Wake, by Amberwolf
Reactions: @gromike
Reactions: @Guerney
Does that mean you like the post, or the music?
Also, @Guerney, do the newer songs have the same low-frequency problem as you found in the previous ones?
Since I cannot identify the problem at my end, I'm just wondering if it's something caused by the instruments or sounds used in the previous ones (which are not in the new ones), or if it's in the effects, or some other part of the process.
Reactions: @Guerney
Does that mean you like the post, or the music?
Also, @Guerney, do the newer songs have the same low-frequency problem as you found in the previous ones?
Since I cannot identify the problem at my end, I'm just wondering if it's something caused by the instruments or sounds used in the previous ones (which are not in the new ones), or if it's in the effects, or some other part of the process.
Gareki:
https://amberwolf.bandcamp.com/track/gareki
I have tried to make something to convey the sadness, the disheartening dismay, of seeing your home destroyed, the things you live(d) with treated like junk and garbage. It's a feeling I know very very well, from the housefire a little over a decade ago.
I am still figuring that out (how to express it), so this is the beginning of that.
I am hoping that those that listen will tell me if it induced any emotion, and if so, what. If it isn't doing what I wanted it to, I can work on it more to try to fix that.
Uses the Z3TA2+ synth for the bassline, Rapture for the CS80 synthhighs, driven by MIDI I either drew or played in. A lot of audio bits using Functionloops vocals, BlackOctopus' Funky Indian Rhythms and Ghosthack's Shymer for most of the percussion bits (though there's a lot of snipping and shifting of the pieces down to individual hits). I have also used some bits from the "ultimate composer bundle" from HollywoodAudioDesign.
EDIT:
01-10-25: 010625 000001 300038m - assorted edits, added elven phrase " This garden you created, this forest you gave life to, is, and always will, be my home." It was the closest thing I could find in what i had to some of what I wanted to say. I have enough words in the assorted phrases in there that I could say more if I knew what all of the individual words actually mean, and the grammar to be used with them. (if I find time I suppose I could research that, as I'm sure it's documented out there somewhere, assumign these phrases follow those rules--presumably it's from Tolkien's Elven, but I don't know, the Shymer documentation doesn't say).
https://amberwolf.bandcamp.com/track/gareki
Previous version without spoken phrase here
01-10-25: 010625 000001 200034j -- assorted edits
https://www.soundclick.com/amberwolf/?more=14920982
EDIT: 01-11-25: 010625 000001 300045p -- converted bass arp to midi notes and edited them to fit the rest of the song parts, modulated delay on bass to cut out unncessary repeats. Added pitch bends and modulation to CS80 lead to better match vocals. Many small timing and volume edits
EDIT: 01-11-25: 010625 000001 300051q - Previous version failed to bend notes correctly in the export (worked fine in realtime listening) so had to do a convoluted mess of a fix to render the CS80 to audio and get this version. :/
EDIT: 01-12-25: 010625 000001 300064s -- Fixed some wierd stuff that happened to several percussion sections' fades (somehow they got "faded in" as clip fades with the end of the fade several clip lengths *past the end of the clip*. I've seen this now and then over the years and I still have no idea what action I have done that causes this; it should be impossible. Many small timing and volume edits to percussion, bass, etc., to tighten up sections. Replaced a bunch of percussion right after and within the first break to change the impact of the section.
EDIT: 01-12-25: 010625 000001 300072v - more timing edits, cleared out a bunch of bass arp notes to make room for things to breathe.
EDIT: 01-16-25: 010625 000001 300082x - Fixed more of the percussion between the intro and the break-in. Added strings accents to bassline and counterpoint to short synth "solo" break near the end, using SI Strings Detache.
EDIT 01-17-25: 010625 000001 300090z -Sculpted percussive transitions, cleared certain percussions to make room for things to breathe.
https://amberwolf.bandcamp.com/track/gareki
I have tried to make something to convey the sadness, the disheartening dismay, of seeing your home destroyed, the things you live(d) with treated like junk and garbage. It's a feeling I know very very well, from the housefire a little over a decade ago.
I am still figuring that out (how to express it), so this is the beginning of that.
I am hoping that those that listen will tell me if it induced any emotion, and if so, what. If it isn't doing what I wanted it to, I can work on it more to try to fix that.
Uses the Z3TA2+ synth for the bassline, Rapture for the CS80 synthhighs, driven by MIDI I either drew or played in. A lot of audio bits using Functionloops vocals, BlackOctopus' Funky Indian Rhythms and Ghosthack's Shymer for most of the percussion bits (though there's a lot of snipping and shifting of the pieces down to individual hits). I have also used some bits from the "ultimate composer bundle" from HollywoodAudioDesign.
EDIT:
01-10-25: 010625 000001 300038m - assorted edits, added elven phrase " This garden you created, this forest you gave life to, is, and always will, be my home." It was the closest thing I could find in what i had to some of what I wanted to say. I have enough words in the assorted phrases in there that I could say more if I knew what all of the individual words actually mean, and the grammar to be used with them. (if I find time I suppose I could research that, as I'm sure it's documented out there somewhere, assumign these phrases follow those rules--presumably it's from Tolkien's Elven, but I don't know, the Shymer documentation doesn't say).
https://amberwolf.bandcamp.com/track/gareki
Previous version without spoken phrase here
01-10-25: 010625 000001 200034j -- assorted edits
https://www.soundclick.com/amberwolf/?more=14920982
EDIT: 01-11-25: 010625 000001 300045p -- converted bass arp to midi notes and edited them to fit the rest of the song parts, modulated delay on bass to cut out unncessary repeats. Added pitch bends and modulation to CS80 lead to better match vocals. Many small timing and volume edits
EDIT: 01-11-25: 010625 000001 300051q - Previous version failed to bend notes correctly in the export (worked fine in realtime listening) so had to do a convoluted mess of a fix to render the CS80 to audio and get this version. :/
EDIT: 01-12-25: 010625 000001 300064s -- Fixed some wierd stuff that happened to several percussion sections' fades (somehow they got "faded in" as clip fades with the end of the fade several clip lengths *past the end of the clip*. I've seen this now and then over the years and I still have no idea what action I have done that causes this; it should be impossible. Many small timing and volume edits to percussion, bass, etc., to tighten up sections. Replaced a bunch of percussion right after and within the first break to change the impact of the section.
EDIT: 01-12-25: 010625 000001 300072v - more timing edits, cleared out a bunch of bass arp notes to make room for things to breathe.
EDIT: 01-16-25: 010625 000001 300082x - Fixed more of the percussion between the intro and the break-in. Added strings accents to bassline and counterpoint to short synth "solo" break near the end, using SI Strings Detache.
EDIT 01-17-25: 010625 000001 300090z -Sculpted percussive transitions, cleared certain percussions to make room for things to breathe.
Last edited:
Guerney
⚡ Regular
- Joined
- Dec 28, 2021
- Messages
- 140
I suspect it originates from the vocal samples, but haven't inspected the spectra of the new tracks. Synths, samples, compression, multiband compression, reverb - wherever superfluous low frequencies originate, it must be eliminated without mercy. Best removed at every stage, rather than just at the end.
EDIT - A little later: Fabfilter is on one of my Windows XP music computers, had to visit the Internet Archive to grab HyperCam 2. By way of illustration without infringing copyright, this is...
AAARRGGGHHH!!!! IT'S EVERYWHERE!!!! WHY HAVE YOU DONE THIS AGAIN AMBERWOLF?!!?!?
I must have Elvish/Elvis ears or something.
It's a common phenomenon, easy to fix. Rid yourself of this troublesome low-end guff! The rest of it looks and sounds ok to my ears. I'll listen with Sennheiser HD600 later, I'm currently using wireless (RF) Sennheiser Rs130.
Last edited:
In each track and each bus fed by them I've tried sticking series of SonitusEQ's setup to take out everything down there in the sub-audible 20hz area at -18db per eq, can't see any difference in the spectrum according to SPAN. Don't have any hardware to play back those frequencies so can't actually tell that way.
I tried some other EQs I have but none of them have a high enough Q to notch out the low end that way without taking out stuff that needs to be there (bassline, percussion, etc).
I'll have to render out a version with these and without, then use the spectrum view in Audacity and see if I can see a difference there. If not, then I'll try the only other thing I think I have with a realtime spectrum view is Smooth Operator by Baby Audio. I can't use it in SONAR but I can use it in Audacity, I think. Or stick it in VSTHost and virtual-cable SONAR or Audacity thru it there, and see what it shows me.
If I can at least see a difference when trying different EQs in the sub end, I can experiment to see which works without taking out stuff I need.
Once I can find a way to see the problem, then find a way to notch it out, then I can apply that as a general filter in my bus templates (and tracks if needed) and hopefully not have to worry about it aterward.
I tried some other EQs I have but none of them have a high enough Q to notch out the low end that way without taking out stuff that needs to be there (bassline, percussion, etc).
I'll have to render out a version with these and without, then use the spectrum view in Audacity and see if I can see a difference there. If not, then I'll try the only other thing I think I have with a realtime spectrum view is Smooth Operator by Baby Audio. I can't use it in SONAR but I can use it in Audacity, I think. Or stick it in VSTHost and virtual-cable SONAR or Audacity thru it there, and see what it shows me.
If I can at least see a difference when trying different EQs in the sub end, I can experiment to see which works without taking out stuff I need.
Once I can find a way to see the problem, then find a way to notch it out, then I can apply that as a general filter in my bus templates (and tracks if needed) and hopefully not have to worry about it aterward.
New version of Gareki.
(no eq changes in this one, is from before those experiments)
1-20-25: 010625 000001 400096A
- intro timing tweaks; added mournful wolf howl accents under vocals
https://amberwolf.bandcamp.com/track/gareki
The wolf howls take a lot of time to add, becuase I ahve to stretch and pitch-manipulate and slice them and fade them together so they follow the vocals enough to not clash but differ enough to add something to them. Still tweaking by ear to try to get mournful sounds that don't interfere with the vocal; i will probably have to also alter the vocals.
(no eq changes in this one, is from before those experiments)
1-20-25: 010625 000001 400096A
- intro timing tweaks; added mournful wolf howl accents under vocals
https://amberwolf.bandcamp.com/track/gareki
The wolf howls take a lot of time to add, becuase I ahve to stretch and pitch-manipulate and slice them and fade them together so they follow the vocals enough to not clash but differ enough to add something to them. Still tweaking by ear to try to get mournful sounds that don't interfere with the vocal; i will probably have to also alter the vocals.
Last edited:
Almost forgot, the barebones just barely started version of Gareki Part 2. For the moment it still has the same vocals as Part 1, until I work out all the changes to come.
It's worth listening to, even if you just think of it as an alternate version instead of a second part.
011925 000001 000007 gareki pt2
https://www.soundclick.com/amberwolf/?more=14929045
It's worth listening to, even if you just think of it as an alternate version instead of a second part.
011925 000001 000007 gareki pt2
https://www.soundclick.com/amberwolf/?more=14929045
Similar threads
