Postcards From Nowhere

I found the blue skies hiding behind the grey
They'd always been there but I didn't see 'till today
Couldn't figure why couldn't see the sky

      Postcards from nowhere were my yesterdays
      I've thrown them away
      What the future brings could be anything
      Tomorrow is the turning of the page

Feels like forever trying to find the way
Drifting through darkness, searching for better days
Couldn't turn around, looking for higher ground

      Postcards from nowhere were my yesterdays
      Dark and stormy skies have blown away
      What the future brings could be anything
      Tomorrow is the turning of the page

The dream is near
It's almost hear

      Postcards from nowhere were my yesterdays
      Dark and stormy skies have blown away
      What the future brings could be anything
      Tomorrow is the turning of the page
      So long, old days
      Tomorrow is the turning of the page


The Instruments
Bass 1397:Bass, Electric, Soul70sSync Ev16 100
Electric Piano: 2194:Piano, ElectricVintage, Rhythm CoolJazzPop Ev 085
Guitar 1: 2448:Guitar, Electric, Background, PopDreamyBrent Ev 085
Guitar 2: ~1435:Guitar, Electric, Rhythm JazzFunkGroovinChords Ev16 110
Guitar 3: 1401: Guitar, Electric, RhythmSoul70sBrightA-B Ev16 100
AcPiano: Piano, Acoustic, Rhythm Soul70sA-B Ev16 100 (Probably)
Alto Sax: Piano, Acoustic, Rhythm Soul70sA-B Ev16 100
Drums: Addictive Drums (Fairfax Vol 2., Faxer 2)
Ac. Guitar: Indiginus Renegade Acoustic (DustyWind)
Lead Vocal: Synthesizer V (Emily Forte AI)
Backing Vocals: Synthesizer V (Emily Forte AI, ANRI AI)


The Effects Chain
Master Buss: Ozone Imager, Lurssen Mastering Console (Americana Loose - More Glue)
Vocal: Make Me Scream, Greg Wells VoiceCentric
Vocal Room: Sunset Sound Studio (Iso Booth, Studio 3), Sunset Sound Studio (Plate Spring, Plate 1)
Sax: EZMix (140 Plate)
Harmony Room: Raum (Grounded)
Bass: Make Me Scream, ReaEq (Basic 11 Band Rock Bass), RBass Mono
El. Piano: Trackspacer, Finisher Retro (Dreamer), Finisher Retro (Suitcase Panner)
Ac. Piano: EZMix2 (Piano2)


The Short Version
This is still a work in progress, so feel free to comment on whatever.

Thanks for listening!

Notes on the Second Mix
Mark Hayes pointed out that the vocals were a bit flat and lacked energy. So I've bumped the key up 7 semitones, and changed the tempo from 100 bpm to 105 bpm.

I also made a small change to the lyric, and no doubt will continue to do some further tweaking, especially to balance the harmony against the melody.


The Blah Blah Blah Part - Feel free to skip it.
This was built around the demo song METRO. I changed a bunch of chords, some of which I got a bit too creative with.

The minor6 chord and the guitars gave it a bit more of a surfer tune feel than I expected. I decided to add a modulation to this, as well as an extended ending.

I decided I should use a bridge instead having three verse/choruses in a row. Little did I know how much effort the sax solo was going to take.

Some day I should write a song with a chorus that's a simple repetition, so it's easier to write and more catchy. Today wasn't that day.

After creating a backing track, I imported it into SynthesizerV and worked out a melody. Unfortunately, the chorus sounded a lot like the verse, and both the verse and chorus were a bit bland. Even worse, the ending of the chorus sounded a lot another song I'd written. But I figured I'd deal with it later and just plow ahead.

The initial plan was to use the new ANRI voicebank that had just been released, and I started with the ANRI voice. But I didn't think it was as good as the Eleanor Forte voice. I tweaked the version to a "male" version, and decided to write to that voice. (Obviously I eventually changed it back).

I wrote dummy lyrics - nonsense verse - to fill in the syllables of the song to help me get started on the lyric. It didn't really lead to anything, but it got me something that could be fixed, which is better than a blank sheet. I went looking through some ideas I'd written while working on other songs, and found an old lyric that I could - with a bit of rewriting - fit into the beginning of the song. I also found the "postcards from nowhere" phrase in my notes, which I'm sure I filched from someone else's lyrics.

Some "fill in the blank" work, and I had a chorus.

At this point, the similarity between this and another song I'd written became really apparent, so it was time to move notes around to make it "legally distinct", so to speak. The result sounded a lot more like real chorus, so I had my daughter listen to the "before" and "after" versions.

She pulled no punches - the first version is "flat", "boring", and "all sounds the same". The second was a great improvement. Not necessarily good, just an improvement.

Yeesh - perhaps I've raised a bit too much like me.

Since there wasn't enough room to develop much of a story, I figured to go ahead with the instrumental. A few lines from the dummy lyrics were salvaged and stuck after the instrumental.

Then I went through the melody and did a lot of edits. There were still a few bits that clashed with the harmony, and eventually I decided to shift back to Eleanor "natural" voice. The song was a bit low for the voice, but... onward anyway!

As the Eleanor voice isn't especially powerful, when an advert for the "Make Me Scream" VST showed up in my email on sale, I decided to give it a try. It was a saturation/distortion effect. It certainly doesn't turn the vocal into a scream, but it does give it a bit of edge. I also used in on the bass, because... well, I did pay for it.

Writing harmony is the fun bit. I tried to make the harmony differ on each verse, to provide a bit more variety.

Then I went back to BiaB to create the sax solo. The new Multi-Riff feature proved helpful, but it still required a whole lot of cutting and pasting. It just didn't seem to want to create a cohesive solo. The key change in the middle of the solo probably didn't help. I spent a lot of time on the solo, and am still not totally happy with it.

The drums weren't working as well as I wanted, to I pulled out the UJAM "Deep" drumkit and started looking for some MIDI that I could build on. I found some patterns I liked, but ended up doing a lot of editing to them. I ended up using Addictive Drums, and putting far more time into working on the drum track than I expected.

I really wanted to have an arpeggiated guitar in the back of the first verse, but again... nothing in BiaB seemed to match. So I got out the Indiginus "Rengade Acoustic" and played in the part.

The drum track was still lacking energy in the final chorus, so I went back and added a ride cymbal.

There are still a few lyrics that I should probably tighten up - "couldn't" shows up too much, for example - and the intro could use some work.

But I figured it's time to release it and get some feedback.

Hello David
once again a great song
I like the background vocals very much
The sax solo is very nice
I don't know if I ever wrote this, but what will be possible in 10 or 20 years?
I wonder if you will even be able to hear a difference from a human singer then?
Is the basis for this voice actually from a real singer or is it all completely artificially created ?

Merry Christmas and a happy and healthy new year 2022

Many greetings
Andi ;-)

Dave,

once again: MOST ASTOUNDING. The song comes along so easily and it fits that voice so well. I really liked the the melody as well as the backing track. And by Jove: It sounds awsome.

You're really putting some effort in your songs...

Merry Christmas and all the best for 2022.

Stefan

Your way with a complex melody is impressive, sir, as is the arrangement! Fun and engaging.

In terms of the bots:

I really like the interaction among the voices, it's well-written and it sounds good.

The lead singer, though, makes me uncomfortable. She sounds "low energy" to the point of indifference or outright depression. As usual, I can't explain where this feeling comes from, but it is strong as I listen now. She sounds like she wants to put down the mic and go home. I trust you know I do not mean that in a way that is critical of you, even though that would be a pretty unkind description of an actual person's singing. =8^)

Not to overdo the "uncanny valley" thing, but I do suspect that might be the scene here. Make the singer just a little more robotic-sounding, and the "depressed" affect would totally disappear. Unfortunately, the implication would seem to be that as bots continue to improve and approach humanity asymptotically, they'll sound more and more disturbing.

And with that cheery thought, winter (or summer) 2021-2022 is upon us!

Amazing work, I especially liked how the BGV played off the main vocal. I'm really impressed. I'm another guy who can't sing so this tech gives us some hope. From what I'm hearing, really good synthetic vocals will come along much sooner than realistically human robots.

Hi, Andi.

Thanks! I appreciate you taking the time to listen and post.


Well, with synthetic instruments such as a saxophone or violin, while an average person may not be able to tell, a professional musician who plays the instrument can almost always tell.

But this is partly because the people who write the music may have the instrument do things that it can't do. If you stay in the range of what it can do, I think we'll cross that gap sooner than people might think.


This is created by having an real singer perform all the phonemes, and then using a neural network learn how that singer sounds.


Thanks, Andi. The same to you!

Thanks, Stefan!


And a Merry Christmas to you. I hope 2022 is a great year for you and Maha.

Hi, Mark.

Perhaps a bit too complex at times.


That's my favorite part of the writing, although I worry that I overdo it more than a little.


You're absolutely right.

Although I really like the low register, it is, as you said, "low energy", and so doesn't fit the song well. Plus, I've got a number of "swoops" where the vocal slides from one note to the next, and that adds to the sound.

So I've gone back and bumped the song to a higher key, and increased the tempo slightly. I wanted to keep the laid-back feel, but it was just a touch too slow.

I probably need to go back and match the background vocals a bit better. I'm not sure I want to pitch them up too high on the first verse, but I haven't had a chance to try it out much.


I'm not sure how much weight to give the "uncanny valley". When that term was being bantered around, people were mainly talking about CGI. We're good at detecting what's wrong (for example, dead eyes in "The Polar Express").

It turned out that a lot of it had to do with poor animation, and I think that's the case here as well.


Yes it is! I figured I'd try to get one last song in under the wire, and I'm sort of amazed that it finally came together.

The best of the season and New Year to you!

Yeah, the background vocals are the most fun part to write.

I'm sure I'm overdoing it, but it makes me happy. That's the important thing!

While it might not be good enough to do a convincing performance, I think it's good enough to put together a demo and work out lyrics and parts.

I certainly wouldn't have been able to do as many songs this year if I didn't have it.

Then it's just a matter of finding someone good on Fivver and spending real money to have them sing it. Or just moving on and writing another song...


Yeah, there are a lot more problems to solve with an android. With vocals, you only to to be able to generate audio.

Thanks for stopping and listening!

I suspect that the energy of the backing will, for sometime yet, be damped by the synth vocal.
I listened passed the vocal to the backing and it's really nice and skips along happily.
The Bing Crosby Effect is significant though.
Lovely melody by the way.

Please, what is this?

Googling only got me "the Bing Crosby effect of stopping me in my tracks and helping me to see more clearly that it’s a losing battle to pursue lasting satisfaction in this life" but I doubt that's what you meant.

I always find your songs to be enjoyable. I think the lyrics match the performers perfectly. Well crafted arrangement. I'm sure it is a lot of work to get your vocalists and the band together in one place but you do it nicely. Good one.

As we said at SC, "An excellent showcase of amazing technology!"

Thanks for the fx info ... mercifully it did not result in us screaming.

We much liked the lyric, melody and the counterpoint vocals.

Reading your post (all of it ), perusing the tracks, and trying to digest your use of Eleanor is overwhelming.
It is hard to comprehend how much thought and DAW time must go into a production like this.

You have the patience of JOB and to your credit the results indicate time well spent given the creative outcome.

Kudos and we look forward to the next one.

Happy Holidays,

J&B

I give the concept weight here because it makes sense and its apparent effect on me really can be really strong.

Remember Macintalk, the "drunken Scandinavian in an oil drum"? He didn't provoke any sort of emotional response, other than yucks. He was a funny robot voice.

Likewise, I think, for decades of synthesized speech.

But now, robot voices have gotten to the point where they're beginning to enter the Zone of Indistinguishability, and they can sound almost but not quite like humans, and for me, that's where spooky, quasi-emotional phenomena start intruding.

Maybe it's like flying a supersonic plane (he wrote, as if he knows what flying a supersonic plane is like.) As you approach the Speed of Sound, vibration and turbulence increase until pow you pass through Mach 1 and things calm down.

You and Eleanor are Pushing the Envelope. Buckle up!

Hi, Ray.

I know what you're talking about. I appreciate people's forbearance with listening to songs being sung by robots - I know it's not everyone's cup of tea.



Thanks! I'm always happy to hear that.


This is the first time I've heard this term. I knew that Bing was a musical innovator, and his investments into recording technologies really helped increase the sound quality of recordings.

Thanks for stopping to listen and comment!

Thanks!

One of the advantages of working from the melody to the lyrics instead of the other way around, is that the sound of the performance helps guide the lyric.

Getting things to gel isn't as hard as all the long list of plugins implies. I just happen to have a bunch of effects that I bought because I like the sound, and I thought they would be a good match for the sound.

I appreciate the positive feedback!

That includes BiaB, of course.


Heh. I figure you're probably the only people who read that part.


Thanks! Always appreciated, especially coming from you!


I try to hide a few tidbits in there to make it enjoyable.

But I really should pare those missives down a bit. I assume that a lot of people do the same level of work on their songs, they just don't bother writing it all down.


Writing excessively long laments seems to be something else I've got in common with Job.


Thanks, and Happy Holidays to you two as well!

You're talking about the version of Software Automatic Mouth that was ported to the Mac. I'm well aware of the program, as I've read and documented the disassembled source code, and written a version in Lua that can sing.

The later versions of SAM are very intelligible and even capable of singing fairly well.. I'd gotten a hold of the company a number of years ago in hopes of convincing them that perhaps using them with BiaB (I've got some code that will convert between MusicXML and other formats). But that never panned out.


Much like how replicators in The Hitchhiker's Guide to the Galaxy can create something that's almost, but not quite, like a cup of tea.

For me, it's enough that it can demo the song. The important bit - for me - is that I can get a level of expressiveness out of it that wasn't there before.

But compare it to an actual person singing, and it's like day and night.

David,

This is excellent! I am completely fascinated by this technology and your approach. I hear tremendous evolution in each project.

You are leading the way on this. Great stuff!

Bob

Interesting, I did not know about SAM.

https://en.wikipedia.org/wiki/Software_Automatic_Mouth

This guy – https://simulationcorner.net/index.php?page=sam
has a demo, I know not from when – https://simulationcorner.net/SAM/sing.wav

Hendrix did it better, but hey.

David,

This sounds really good! I love that jazzy/pop sound! I'm not familiar with the use of those synthesized vocals but they sound pretty good, especially the bgvox! Great tune! Thanks, Torrey

Hi, Bob.

Thanks! I think I'm getting better at using it, and it's getting improved through regular releases.

One thing I hadn't expected was how the voice quality of the singer would impact the songs. I know it's obvious in retrospect, but - for example - the soft voice has a different "attitude" than a harder voice, so a hard rock song wouldn't work.

There's a new SOLARIA voice being released next month. It'll be interesting to see what direction that voice take me in.

Glad you had a chance to listen.

David you are a master at synthesizer vocals and this sounds very real to these old ears.

Your best to date IMHO

Hi David,

You must have put a lot of effort in creating this song, with the vocals, the harmonies and the band.
The melody is very lovely, with sweet lyrics. And I must say the vocals sound more human now!
The harmonies are done very well. I could learn from that. So, I am impressed that all your working with these artificial vocals brought you this far. Hats off!

Still, the band is a bit to static and modest (if that is the word here). Maybe you could use some real tracks from BiaB that 'live' a bit more. And also the vocals are also a bit lifeless. They need to sparkle more, but I think that will be a hell of a job to achieve.

As I said, I am impressed!
Happy Newyear,
Hans

Hi, Torrey.

Thanks! I love all those '70s sounds that BiaB has, and I don't think it's possible for me to write a song without using a bunch of major 7th chords.


I'm glad you enjoyed it!

One nice thing about robots is that they don't complain there's a difficult interval to sing. I'm probably forming some bad songwriting habits because of this.

And being able to write background vocals that intertwine with the primary like this is a lot of fun.

Thanks for the positive feedback!

Thanks, Mario!

I suspect that when other people start posting using synthetic vocals, you'll find that it's more the software than me. But I'll take the compliments while I can.


Thanks! Good to hear.

Hi, Hans.

Thanks, I'm glad the song works for you.

I'd like to take credit here on the voice, but I suspect it's mostly because the software has improved, and this song is a better fit for the range of the voice.



There's not a lot of back and forth in pop songs, although there are some exceptions. The Christmas song "Baby, It's Cold Outside" comes to mind.


You're right - there's not as much interplay in the instruments as there could be.

Part of the reason is that I had initially written this in another key, and the guitar part worked with the vocal a bit better. When I transposed it, I lost a bit of that.

I also rely more on rhythm instruments, and don't have as much variety in the arrangement as I could have done. I was a bit more focused on getting it finished than building the arrangement as much as I could.

I suspect that adding a better bass track could help. Maybe I'll see if I can get my son to play one.


Thanks again!

I appreciate the feedback, both positive and negative!

Hi David,

Very nice song and enjoyed the vocals - I can't believe they are synthesized - wow!

Best Regards
Nigel

David,

Very nice song!
Excellent melody and chord progression.
The use of Synthesizer V is very fine as usual.
Enjoyed my listen a lot.

Best regards.

Shigeki Adachi

Hi, Nigel.

Yes, the technology is pretty amazing. It's a bit mind-boggling to go back and compare where it was a decade or two ago.

Thanks for the positive feedback!

Hi, Shigeki.

Thanks, that means a lot coming from you!

Although I stick to the circle of fifths a lot, I try to make the chords a bit interesting, which often leads to some melodies I wouldn't have ordinarily tried.

I'm glad you could get a chance to listen to it.

Hi David ...

Well, I did read "ALL" of your introductory info. Most of it was Greek - or some other foreign language - to me. You've probably read how archaic and simplistic my recording process is.

The lead vocal was quite nice. What I really enjoyed was the counterpoint vocals - they really add to the entire project. Nice sax solo and a terrific mix for the entire song.

At near 75 years old, my ears sometimes lie to me or just ignore everything other folks hear - but .... do I hear a small bit of vibrato in the Eleanor's voice? If I am correct in that assumption, perhaps that is why I think she is sounding more and more human as her vocal history has unfolded. Also, I always love your chord progressions. I love playing around with chord interactions and see how I can come up with something different and interesting but isn't exceptionally jolting.

I admire your persistence, creativity and overall production skills. Impressive!

Wishing you and yours a very Merry Christmas and the best New Year ever ...

Alan

Nice title leads to the right mood which is expressed very well in the slightly positive lyrics, in the verse melody and the instrumentation, I really like the guitars. 70's feel in the band and chording, although I cannot recall the female artist whose music it reminds me. Somehow also reminds me of songs in Japanese anime movies... Interesting to read your production details, quite a lot of engineering, anyway very good work with Eleanor and those counter parts are impressive.

Janne

Hi, Alan.

Well, you are a glutton for punishment!

The right perspective on this is that my process is overly complicated, and you're doing something right.


Yes, the counterpoint is the fun bit. The underlying melody and vocals are hard work, as is the mixing and fine tuning.

But the counterlines - those are fun to do. So it's good to hear they sound good as well.


Yes, there's a lot of expressiveness in the vocal, including vibrato.

This version of Eleanor is an "AI" version, which uses a neural network to generate the voice. There's a slight hit to the vocal quality, but it's generally a bit better in overall expressiveness.

Also, it's been trained on fine pitch detail, so you have an option to "humanize" the singing, which makes it more natural than if it were just randomly jittering the voice.

There's also fairly fine level of detail that can be performed manually, and I've written some script to help me with adding things like progressive vibrato.



Thanks, I appreciate you noticing and mentioning that!

Merry Christmas, and I hope this year brings the best to you and Di.

Hi, Janne.

I like a sad song as much as anyone, and was planning on doing a super-sad sort of thing with lush strings and an english horn.

I was also hoped to do a more "Carpenters" sound with the vocals, but wasn't having a lot of luck with the thickened harmony lines.

But I'm happy with how it turned out, even if it was more upbeat and positive than I'd initially intended. Songs seem to go the way they want to go.


And I'm happy to see the Production Notes didn't bore you to tears.

Thanks for stopping by!

David,

Goodness! it takes a while to read your threads (especially if you've been away or don't have time to get to it for a few days)!!

I do enjoy the fact that you document your thoughts and processes. Interesting reading. Your thoughtful responses to comments, too...

The BAND/tracks sound great. The opening gave me a Steely-Dan-vibe...great company to be in... the mix is tight.

The robot voices ARE impressive (technologically). But...it seems as though "they all sound the same". Not just your use of them, but ALL uses of them. You do a hell of a job getting them to sing with a good cadence. But...they, inevitably, lack true emotion.

You write very cool melodies. However, (I didn't want to say "But..." again ) This one feels like a trying-too-hard moment. The melody "bounces" a lot. Ups and downs - which works at times, then, at times, just feels like it is up-and-downing unnecessarily - which begins to sound forced. There actually needs to be more "sameness" ("even-ness"?) to the melody to make it sound "natural", not "manufactured".

...this is in NO WAY intended to suggest what you are doing is not VERY impressive (all aspects of it). It's VERY impressive. The acoustic sounds wonderful! The harmonies are exquisite. The drums are tasty!

You are a very talented musician and songwriter and producer.


fj

I started out thinking, "just give it time", then tried to imagine what the future might be like.

I succeeded in scaring myself.

Imagine this:

It's 2035. (David might be able to project a little more precisely.)

You initialize a singer by firing it up and interacting with it. You start off with some unscripted casual chit-chat to get both of you comfortable, then move on to singing together.

At first, it sounds like you are chatting with a robot. But as the training progresses, the singer learns from you and emulates you.

Eventually, it sounds like you are singing a duet with another you. Spooky!

Then, the emulation becomes so perfect you can no longer hear the other voice at all. You can only tell you're not alone by watching the meter.

You try to throw it curves, singing off-key or off-beat or with totally inappropriate affect, but it anticipates you perfectly.

Beep beep, your singer is trained.

Man, this sounds really cool!

I love the music. Robot voices will always be that, no matter how much they "improve" I suppose, but I am impressed by what you are able to do with technology for sure!

Merry Christmas!

David,

you are really making progress on Synth5 and the Eleanor/ANRI . I now too own the Pro versions and look forward to learning how to tweak/tune the software. i look forward to following your AI blog too

Hi, floyd jane.

I do tend to write a lot. A lot is re-processing what I'd tried to do with a particular song, and see where I'd hit or missed particular marks. Plus, songwriting and production tends to be a very solitary process, so it's really helpful to talk and get feedback from other songwriters and producers - especially those I admire!

So I appreciate all feedback, both positive and negative. It helps confirm places where I already think I need to improve, and point out places I've entirely missed.

I'm not just bumping my post.


Good to hear! I'm a big fan of Steely Dan, and the Alan Parsons Project, and I suspect I'm more producer than songwriter.

But that also leads to problems when I figure I can fix something via production rather than going back and fixing fundamental things like a hyperactive melody. It ends up pretty, but still fundamentally flawed.


I hear what you're saying. I've spent a long time trying to get my own vocal synthesis program off the ground, and no matter how much improvement I've done, there's always a huge gap between what it does and what a real singer can do.

And I always feel sad when I hear someone who uses strings with a "cheap" MIDI library, because it feels like it just lets the song down. So when it's the vocals, which are the key to the song, it's even more so!

A lot of these songs simply wouldn't have been possible to do without a vocal synthesizer, since my home is currently anything but ideal for vocals. But I've also got a house full of singers (myself included), so I should make more of an effort to get real vocals on my songs in New Year.

In the meantime, thanks for putting up with the robots!


Yeah, it's a "threading the needle" thing for me, and I'm obviously not there yet.

My verses often feel like they're trying to be the chorus. I'm spending too much effort trying to outline the chord changes and create interesting lines, but that's not really the job of the verse.

I think part of the problem is the way that it's created. I'll start with chords, when melody is the real driver of the song. Then I'll build the melody up one phrase at a time, without enough attention to how it matches what's gone before. Then I go back and try to make it more unified, but still focus on the micro level of the melody.

In the future, I'll try to re-use motives, have less ideas, and avoid using so many leaps that sound lovely, but are out of place in the verse.


No, I appreciate the honesty here.

And thanks for mentioning the drums.


And, of course, I appreciate any praise in addition to constructive feedback!

Thanks again. It's always good to hear from you!

There's already "voice cloning" technology, and it only takes less than a minute of training data to create a credible clone of a voice.

Check this list out for examples.

That's not the same, but it shows that the technology is feasible.

In a similar vein, the developers at SynthesizerV just announced they have implemented cross-lingual synthesis. That means that voices that were created in one languages can now sing in another language:



There are already voicebanks which already have "English" versions by singers who aren't native English speakers, but they often have heavy accents, and poor pronunciations of phonemes like /r/. There are still slight accents in the cross-lingual versions, but they much more usable.

Hi, David.

Yeah, I think they're good for demoing the song, but - especially if you're a singer - you're going to quickly hear the difference.

But it's still a useful tool for putting together a demo of a song. Especially when conditions aren't ideal for recording vocals, like when the rain is pouring down on the flat roof above, and someone's playing video games on the couch less than ten feet away.

Merry Christmas to you and your loved ones!

Hi, Bob.

You can always send me a PM, if you've got any questions.

As far as my own vocal synthesis project goes, it's been on hold for the last month, but I hope to get back to it next year. One of the problems is that building a good vocal database is a whole lot of work. It requires recording vocals, manually tagging the phonemes, and lots of fine-tuning.

So if I can't get the quality level high enough this time around, I think it'll just make more sense just to quit. SynthesizerV basically does what I need it to do, and I have a feeling that better Open Source vocal synthesis projects will be around the corner.

But I'll have one more go at it, just because I'm stubborn.

Wow..! Interesting, cool technology. I've listened to several of your tunes, and I'm very impressed with your creativity, and talent.

I really, really like this!!! Some wonderful and unusual musical changes, and a very cool lyric!! And, although I'm not a fan of the vocaloid thing generally, I think it really works here....and the whole mix is great!! NICE!!! Take care. Greg

Hi David, thanks for your so detailed answer.
I thought it would be a lot of work to do that song.
And I wasn't negative at all, only it takes more to get the human touch right.
And that's been your quest for some time.
I like that.

Hans