Stimulated by DCuny's post about voice synthesis I started experimenting
with Sinsy and wrote a little song. (Thank you once again for the motivation).
Suitable for the subject I used Trance style and produced in FL Studio.

RealTracks are:
RT1771_Guitar, Electric, Rhythm JazzFunkMovinMuted Ev16 130
RT1444_Sax, Alto, Soloist JazzFunkGroovin Ev16 110
RT1273_Guitar, Electric, Soloist PopUplift Ev16 120

As I happened to get a trial version of Vocaloid's Megpoid (Gumi) I made a second
version with that voice.
I personally prefer Sinsy because it sounds really artificial. Gumi is a little
too 'human'.

http://soundcloud.com/boehmusic/artificial-voices-sinsy
http://soundcloud.com/boehmusic/artificial-voices-gumi-version

I suggest good speakers or better headphones to get the
subsonics and panning effects.

Guenter

Hi Guenter,

Very impressive! I didn't catch all the words on either version but the aural texture of a human voice is clearly recognizable on both. Personally, I preferred the Gumi version.

Once more I'm amazed by what you do!

All the best,
Noel

Guenter,

As exceptional and interesting as everything else you have done. The music is beautiful and thoroughly enjoyable. The instrumentation - great - as always - the sax and guitar are quite tasteful and... tasty.

I actually preferred the gumi version, because... I could understand some words (quite well). Some I could not. That made it like some secret message that I might figure out if I listen enough.... intriguing. (the sinsy, I could not understand much...)

Cool stuff. I am a fan. Uhg... Guenter... good...

floyd

Thanks for posting. One question, with these voice synths, how do you control the pronunciation, for example, if you want them to say amen, how do you control whether it is ahh-men or ayy-men ?

Hi, Guenter !


I was mighty impressed with your artificial
voices indeed !:)Soon we old singers will be
completely replaced by artificial ones (LOL) !

Cheers
Dani

Guenter,

in my opinion this is an excellent piece of work even before the artificial voices enter the equation!

You have a real knack for assembling interesting instrumental parts into one cohesive sound quilt. What's your musical background? Surely you aren't just a hobbyist. Hobbyists tend to make nice songs that sound like other nice songs they've heard before; but all of your creations have an air of originality that I don't often see/hear on hobbyist sites.

Very nice work! It would be an understatement to say that I enjoy your songs.

Thank you all for your nice comments.
@Noel and Floyd: I think it's not important to understand all the words. But here are the "lyrics":
"Looking up, looking down
illusions all around.
Artificial voices
Supernatural sound
I'm flying to the sky
I'm soaring way up high"
@Peter: In Vocaloid there is an English Dictionary with a phonetic transcription which you can customize.
For example: I wanted "natural" to be pronounced as 2 syllables. So I made up the words
"natch" n@tS and "ral" r{l.
So amen could start with eI or Q@.
In Sinsy you must experiment with the writing of words. I could try it with "armen".
@Dani: Don't bother. We won't experience that.
@Pat: Yes, I'm a hobbyist. I started learning the piano at the age of ten, with sixteen the bass and later drums and saxophone.
For twenty years I played in a band (mostly in American officers' clubs), and for another twenty years I ran a sound studio.
But that was all hobby beside my actual job of being a teacher.
(Although one of my subjects was Music for about twenty years).
Now I am retired and only want to be creative. That's why I experiment a lot. I just love the process of creating.
And it's great fun to find out that some people like my stuff.

Guenter

Carl...

Swingin' Electronic Music...

Stationed in Ansbach, 1957-58, I played gigs around the 7th Army and for Fasching.

I've played synth guitar (Roland GR-303) and have enjoyed my toys...some listed below,

for many years.....but it must swing, right???

Okay, we know the Real Tracks--but how did you do that bass line? This track is comparable to the best of Stevie Wonder without being derivative in the least. I can think of no higher praise. I bow before Meister Boehm.

Srsly, this thing rocks! I love the electronic setting, and the voice just fills out the palette. Mehr, bitte.

R.

Hi Guenter
what an amazing piece of work!
I'm not into electronic music but this is AMAZING and may have changed my mind!
I dont think the voices HAVE to make sense - a bit like Carl Jenkins with his "Adiemus" suite.
You manage to blend in so much into a cohesive whole.
cheers
ian
BTW - did I say it was Amazing?

Thanks GDaddy, Danny, Ryszard, Pat and Ian
for the nice words.
@GDaddy: It's nice you call me Carl. But there has never been anybody named Carl in my family. You're right, it must swing (or rock).
@Ryszard: That's too much. You make me blush. Stevie Wonder wouldn't like that.
@Pat: Ok. Call me a semipro. Thanks a lot for your compliments.

Guenter

Like Noel said, the Gummi version is a bit nicer, because it draws less attention to itself. When I hear something spoken, I automatically want to try understanding it. With the Gummi version, I can just zone out with the music.

And as Pat said, the bass like really makes the song.

Are there still any demo versions of Vocaloid available? It would be useful for my not-so-secret coding project.

Guenter
A very impressive piece.

Ryszard -

Thanks for the tip on boosting the bass line in Estate. I did it, and, sure enough, it sounds better. My guess is you're a bass player, because you asked exactly the question of Guenter that I wanted to. Namely, where the ,,, did he get that superkickin' bass line? Is the answer somewhere in this list of responses?

Aleck Rand

WOW!

Guenter, I've been looking in this list for the source of the bass lines. Who - WHAT- did that?
Du bist ein Monster mein Freund! Und damit meine ich nicht einen Mann in einem Gummianzug, die du siehst in den Filmen! Du bist WIRKLICH beängstigend.

Aleck Rand

Thanks David, Joanne and Aleck for your nice comments.

@David: Here you are: https://www2.ssw.co.jp/download/dl_trial_en.aspx
Where do I get information about your coding project?
@Aleck: Here is the source for the bass line:
Who --> me
What --> Sytrus synth (in FL Studio)
I have been a bass player for 50 years (first bass guitars and for some years on keyboard) and that bass track wasn't that important for me. I'm glad so many people like it.

Guenter

Hi, Guenter.

Thanks for the link!

My goal is create a free Vocaloid-like program that can sing reasonably well in English.

I'm focusing on the rendering engine, rather than the GUI. For one thing, I think Sinsy shows that it's possible to get reasonably good output if there's some built-in intelligence about phrasing and such.

There's already a free editor for Vocaloid/Utau written in Java called Cadencii, so there's no need to reinvent that wheel.

I've done some proof-of-concept work. The pitch detection and pitch shifting code seem to work reasonably well. Here's an example. You can ignore the "buzzy" sound - that's part of the original recording. The important thing is that it (more or less) preserves the character of the sound without introducing the "chipmunk effect".

I'm creating a sort of "melodic backbone" of frequency information that includes preparation, portamento, overshoot and vibrato, and while it's pretty rough, it shows promise. Right now, it only renders the melody as a sine wave. Here's an example.

I'm currently in the process of gluing these parts together into a simple rendering engine that can take phonemes sung one one pitch, and follow that melodic curve. Once that's done, I'll can have it start singing "La, la, la".

I've also experimented with cross-fading phonemes together, both with simple time-based crossfades and more complex harmonic-based crossfades (like in Vocaloid). I can't hear much difference between the two. When I add that, I should be able to render simple melodies like "do-be-do-be-do".

If I can get it that far, I'll need to start building a sung database of phonemes. I'm not looking forward to that, because English requires a ton of phonemes. I expect to record blocks of blends in VCV format, like "ahtah ehteh ihtih ohtoh uhtuh aytay eetee ietie ohtoh ...", much like UTAU does. Each of the blends can be used for VCV blends ("...ahtah..."), VC endings ("...aht") and CV beginnings ("tah...").

It's being coded in Java mainly because I'm comfortable coding in it. I suspect it would be fairly straight forward to convert to C or C# when it's done.

I'd considered trying to do this with the UTAU engine, but there's no documentation on it that I can find. Also, it seems to require Japanese localization, and that's a bit of a pain.

If you've got any additional questions, let me know, or shoot me an email - I PM'd you my email address.

This is a terrific composition. Well produced.

I love the cross-pollination of soft synths and Real Tracks. BIAB is so unique and makes it so fun add real samples/phrases to electronic music as you can literally write out the progression and the real track 'musicians' will follow suit. Extract those...and you have a well played...pre-played track to add and manipulate. The RT Sax really worked well with this. Soft and yet driving. Great work!~