Slashdot Mirror
Yamaha Releases Singing Synthesis Software
loopdloop writes "The world's first singing synthesis software, Vocaloid, was released by Yamaha this month at the Los Angeles NAMM show. Simply type in the lyrics and notate the vocal expressions to create a completely computer-generated singer. There are also audio demos of the product available." Update: 01/26 21:14 GMT by S : An earlier NYT-authored preview of this software has also been covered on Slashdot.
Great, now we have to listen to computers do karaoke too?
I guess cheating at Karaoke Revolution is a cynch now, though. Just hold the mic up to your speaker. And a plus - I don't have to hear people wailing in the next room! Well, not as much, anyway.
Liberty in Our Lifetime - http://www.freeme.org/
"I was going to do that?"
This was something I was really interested in when choosing a college major, and thought that I'd get into EE CS and do this. Somehow, I've found myself coding web applications instead.
I'm glad to see somebody's doing it, but man, I think I took a wrong turn somewhere.
Tweet, tweet.
Vocaloid has been covered on Slashdot before. It is one of the many impressive projects to have at least in part come out of the Music Technology Group at Institut Universitari de L'Audiovisual in Barcelona.
This is one of many impressive Music Technology groups in the world who is kind enough to provide us with open source software such as CLAM. Similarly there are some groups out there doing interesting things. Needless to say, I could link all day...
I am a graduate student in this field
Comment removed based on user account deletion
That's quite amazing. Now we need a computer to write music and songs.
-Tim
Shpongle (trance group) used Vocal Writer in their CD that was released in 1998.
...this is where Britney Spears' talent comes from!
Edward@Tomato - /home/Edward/ man woman
man: no entry for woman in the manual.
"Qua!?"
man, at least Milli Vanilli had singers.
It had multiple voices... and was fun. MC Hawking style.... "mmmmmmmMMMMMMMMM ya"
You talk better than you fool!
Wow, this must make the RIAA's day. An artist who needs absolutly no pay and who really is property...
Since when has this country used intellectual elite as a pejorative term?
Still sounds like shit
If you get rid of the background instruments, the synthetic voice still sounds
...first, dancing robots and now singing computers.
Between Fred Durst and Kid Rock, this sound isn't new. Now, maybe could we replace the actual Fred Durst with a simulated turnip from Yamaha?
It's called an 808 and a 909 and they have been out for a LONG time.
Wait, do I need to take the hook out of my mouth now?
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
Machines can _never_ beat human voice. No matter how much technology advances, it will never happen. We are lucky to have emotions - machines can never become emotional. It's all just fake.
"The world's first singing synthesis software, Vocaloid, was released by Yamaha this month at the Los Angeles NAMM show.
Feh! They might be able to program something that sings better than Britney, but until they integrate it with something like this, Ms Spears' talents will continue to be in demand...
*** Where are we going? And what's with this handbasket?
So Britney Spears would be the organic version of this software?
I guess Britney is going to be out of a job now. We all know that any computer can sing better than her, and since neither of them play an instrument, I guess she's screwed.
Not to mention that the computer is far sexier.
Now there's no need for Britney, Justin, or any of those other canned-pop "vocalists" (term used extremely loosely)...
Now American Idol doesn't even need to find teenie boppers to exploit for cash, they can just use the computer, have something much more talented, and work the thing to death..
I'll take three!
So will we finally get to replace the prime-time T.V. show American Idol??
Quake master!
"Studies have shown that people who eat peanuts live longer than those who do not eat."
Volcaloid Idol! Maybe they can turn that Simon guy into a robot too.
What about using it to give expression to the voice and creating virtual actors. That with CG would result in much cheaper movies :-) No millionarie contracts anymore ...
computer that can drum? we had that 20 years ago, it's called midi
.ASF .ASX .WMA
What's Linux user to do???
I can now see the day when tweens have poster's of their favorite "singer's" Vocaloid profile in XML.
Or maybe take the 0s and 1s of it and make ASCII art to look like Britney.
Now anyone with a computer can be famous!
All you need it tight pants and a sock roll.
The flashbacks are horrendous!
Blame in on the rain...
ender-iii
that the Pop Idol contestants had been puppets for this technology since the show's inception. Dont tell me they really DO sing that bad?
It could be that the purpose of your life is only to serve as a warning to others.
Mod up, it's the truth.
it sounds like this
gotta love the shat.
Uh, that's not a troll, but you apparently missed the joke. Or should I take the hook out of my mouth?
irb(main):001:0>
The logical next step would be a program that would listen to, and enjoy, the music that other computers write and sing.
Think of the time it would free up, and the money it would save - you would never have to buy CDs. *cough* of course, some people have already eliminated that expense.
they can retire Britney Spears and just replace her with a touring Vocaloid. (And maybe some strippers.)
TALENT!?! Please, her only talent is looking good.
Now i can finally get a computer to sing and play the music while i do the beats.
Wow, I didn't know Tony Verderosa posted to Slashdot!
But can it rap?
:P
I might actually hear some talent at my job for once
Steven Hawking is trying to start up a band.
Just listened to these.
While their "Volcaloid" tech is nice, their "Lyricoid" tech needs work.
The same technology was used a few years back by the boy band The Party Posse.
It's all neat and stuff that they've done this, but "the world's first singing synthesis software"? Not by a long shot.
Perhaps next month we will get to see an article by Slashdot on the subject of Mitsubishi releasing the world's first car.
Irritable, left-wing and possibly humorous bumper stickers and t-shirts
Apparently you missed a joke, too. :)
OK, in case you didn't know, the 808 and 909 are REALLY OLD, OBSOLETE (by most people's standards) drum machines.
assert(birth_date<time-86400)
I had the chance to try it out at NAMM and it is VERY difficult to get it to "sing." It can probably be used adequately for backup vocals, but again, it takes a lot of work to get it to sound human. Nevertheless, a step in the right direction.
A blog like any other.
I think my Sound Blaster Pro came with this software. It fit on a floppy disk, and you could make the computer sing whatever you typed in. In fact it also came with a psychiatrist named Dr Sbaitso. Just don't cus at him, he gets offended very easily.
Seriously, I'm sure that this new software is much better. At least I sure hope so...
Interestingly enough, this made me think of something I read in William Gibsons blog a long time ago. I don't know where it is now though. It was about how in the future, people will be able to take a movie or something on their computer, and tell the computer to replace all the actors heads with dog heads for example, and change what they do and say with simple commands. Perhaps this software is the lower level beginning of making that happen, we'd just need some higher-level controls to make it easy for everybody to use.
Buy Steampunk Clothing Online!
I'd rather hear Dr SBAITSO sing.
Remember that? It's the command line computer voice synthesis generator program from Creative which sounds like Dr Hawkins.
Wonder if it's still around...
Trolls dont like to be Flamebait, because they burn so well. Protect our Troll heritage!
I'm relieved after listening to the demos that REAL SINGERS are in no danger of losing work.
Britney Spears, on the other hand...
Since they're so old and obsolete, if you find one in your attic, I'll take it off your hands for a cool $20 USD.
(No disrespect intended).
Sigs are bad for your health.
Damn. So much for musicians and instruments. Long live programmers and the computer! Now, if I can only get ahold of a lisp interface to thesel ibraries, I could take over the music industry from my computer and crush the RIAA!! Muwaha.
Fitter, happier, more productive
Comfortable
Not drinking too much
Regular exercise at the gym
(3 days a week)
Getting on better with your associate employee contemporaries
At ease
Eating well
(No more microwave dinners and saturated fats)
A patient better driver
A safer car
(Baby smiling in back seat)
Sleeping well
(No bad dreams)
No paranoia
Careful to all animals
(Never washing spiders down the plughole)
Keep in contact with old friends
(Enjoy a drink now and then)
Will frequently check credit at (moral) bank (hole in the wall)
Favours for favours
Fond but not in love
Charity standing orders
On Sundays ring road supermarket
(No killing moths or putting boiling water on the ants)
Car wash
(Also on Sundays)
No longer afraid of the dark or midday shadows
Nothing so ridiculously teenage and desperate
Nothing so childish - at a better pace
Slower and more calculated
No chance of escape
Now self-employed
Concerned (but powerless)
An empowered and informed member of society
(Pragmatism not idealism)
Will not cry in public
Less chance of illness
Tires that grip in the wet
(Shot of baby strapped in back seat)
A good memory
Still cries at a good film
Still kisses with saliva
No longer empty and frantic like a cat tied to a stick
That's driven into frozen winter shit
(The ability to laugh at weakness)
Calm
Fitter
Healthier and more productive
A pig in a cage on antibiotics
there's no place like ~
What you don't believe me?
True. Emotions can not be programmed. BTW, do animals have emotions is is it only a human thing?
I can't be the only one that thinks the background vocals sound like Electric Light Orchestra??
There's Sprectrasonic's Symphony of voices, which is recordings of voices singing, (in the song Sonne, by Rammstein). It's not the same thing; it has Oooh's and Ahhh's, but not actual words, although that seems to be the logical next step.
Mod "Overrated" instead of replying "I disagree with you," you coward.
Just like many Yamaha firsts, this one may have been overhyped a bit. This sounds like a real person singing in the way that a synth brass pad sounds like a trumpet. There is no way in hell you would ever even consider that these noises were made by a human being. Yes, I understand that most of the samples are in Japanese and might not sound normal to me anyway. But, even if you listen to the ONE in English alone, it sounds like the Bell Labs female voice, but screechy and obnoxious instead of like a drugged out cigarette smoker after a trachyotemy (sp?).
Webmaster Wanted - Entropic Reactions
How about an app that finds the tablature and lyrics to a song, marks that data up in the Yamaha format, and sends the arrangement to the singthesizer? It's like PostScript for singing. Where's my Perl module?
--
make install -not war
Now we can finally get rid of these whiny musicians, always complaining about "I need to feed my family" and "I'm a professional and should be paid like one." Now all of those unskilled morons can be sent to fill up the thousands of food preparation and customer service jobs that our public school system can't seem to find enough people for.
Sorry about the offtopic (tongue firmly in cheek) rant. You're right. this does sound like a fun toy.
If you think a computer's sexier than Britney, maybe you should change your name to "gay ninja". Faggot.
I can't find a link to an actual demo of it simulating a human voice, but here's a page that documents its use to reproduce the sound of a suling (javanese wooden flute). Does a good job too. I've heard it demo'd with a human voice, and it was pretty good (though the neural net needed additional input - the syllable being sung - obviously).
i'm sure that many of the other academic computer music labs around the world had similar software long before yamaha introduced this package. still cool, though.
No not the first...
The Mac speech thing could be programmed to sing.. I remember doing it maybe 10 frickin' years ago! Just get it to talk asynchronously, and while the speech is running, change the pitch. Easy!
I know of at least VocalWriter which uses this technique and has been out since at least 2000.
Not the "first" by a long shot.
I'm sure Britney's "talent" has absolutely nothing to do with her ability to do vocals, and absolutely everything to do with her abilility to take off her clothes...
"Freedom means freedom for everybody" -- Dick Cheney
Talk is cheap, so if you can't implement your invention, punt! Explain it in detail in public, or among a relevant developer group. Then you at least have a chance of being in on the creation, along your lines of vision, and at least get your world bettered, even if you can't cash in by doing the hard part. The joy of coinvention sure beats the bitterness of "coulda, woulda, shoulda". Most of the open source process is based on that crosspollination and mutual assist. Hell, if we all did this better, maybe the docs would get written *first*, rather than never-quite.
--
make install -not war
Virtual Singer has been a favorite of mine for playing around.
The software is of (lesser) but similar quality to the demos I just checked out for this new engine.
But, it also begs the question: what *really* is the purpose of software designed to replace humans going to be used for besides... replacing humans?
The appeal seems to be to small-time musicians/hobbyists (no money to pay people) and big-time corporations (wanna keep more money for themselves). 8/
Diplomacy is the art of saying, "Nice doggie!" until you can find a rock.
Creative's TextOLE had "singing text" back in 1996 - I had demos on my PC that came with my SoundBlaster 16! You marked-up text with notes and expressions for each phonym and the text-to-speach synth could sing.
It did a great rendition of "Itsy Bitsy Spider". * grin *
The one thing I expected they wouldn't get right, was what they did the best.
When people hold notes, there are natural fluctuations in the tone, nobody can hold a perfect tone without some audible wax or wane.
But you can hear this simulated amazingly if you listen to that one japanese song with the single male "vocalist".
Vonal Declosion
Lyrics: Developers! (repeated infinitely)
Vocal Expression: Steve Ballmer. Cracky voice. Exhausted.
I havent compared the qualities but myriad software wrote Virtual Singer years ago, http://www.myriad-online.com/main.htm they also have alot of other cool music writing programs that thankfully aren't just 'e-jay clones'
...Now where are my brain interface mech-planes?
"Geez, I hope they don't make a computer that can drum."
;)
:) http://www.apple.com/ilife/garageband/
I'm a drummer too... and unfortunately, I think we were the first to get replaced.
Heck now it's got'n so cheep that Apple will let you bust out MIDI or WAV looped percussion for $50bucks
"Things are more moderner than before- bigger, and yet smaller- it's computers-- San Dimas High School football RULES!"
I heard she plays the skin flute and the hairy harmonica.
oh, no! teh singing software! ti is released!!!1111 someone has to catch it!!11 omfg wtf1!11111^^^^
Can you imagine a beowulf choir of these?
Not only that, but we can also get rid of composers and replace them with "trial-and-error" number crunchers. Finally, freedom from the RIAA!
The Ezine Directory
I wonder how long it takes before RIAA gets their grubby little fingers on this bad boy and makes it illegal to type in known lyrics... (copyright infringement!! right? :P)
Nothing disturbs me more than blind loyalism towards some unrealistic and over-idealistic notion of one's nationality.
Ditto. 'Cept you had to spell everything phonetically. Hearing it "mis-pronounce" english was pretty fun.
In the future, I would want to not be isolated from my friends in the Space Station.
i think that it won't be long before synthesized vocals take over a significant chunk of the recording industry. this includes voice-over tracks for things such as advertising and animated films.
S.A.M. on my Commodore could carry a tune just fine, thank you very much!
You kids today with your crazy singing machines. In my day, we had to manually adjust the pitch of our robotic voices, and we liked it!
Maybe someone I saw on Slashdot will finally do that parody of "Bye bye!" ("vi! vi!)
tasks(723) drafts(105) languages(484) examples(29106)
We have computer programs to write words, programs to sing, and programs to rate the singing (Karaoke Revolution). The final step is to write programs that will listen to Britney so we don't have to.
It's bad enough we have Cher and Madonna doing the fake electronic yodel dance, now every dipthong with a little cash will be creating them.
"I'm just here to regulate funkiness."
we no longer need to suffer from listening to nice songs sung by bad singers!
just hook up Vokalazza and start downloading lyrics and some vocal expressions, then we'll have a DIY album.
it probably won't be long until 'authorities' start beating up on such exchange though.
Comment removed based on user account deletion
What about VocalWriter for the Mac. Allegedly, "the world's first music synthesizer that can actually sing your lyrics", it's been around at least 5 years now ...
Alison
"It is a miracle that curiosity survives formal education." - Albert Einstein
Boy, I hope this is a troll.
Someone gets it to sing "Daisy, Daisy" slower and slower until it stops?
Shite!! I pressed the back button and the General Midi Junior High Choir just won't stop playing!!
Yeah, right. Singing synthesis is about as old as speech synthesis itself. Here's an example from 1958 (item 11 on the linked page). By 1961 singing synthesis had become fairly mainstream -- witness the singing HAL in "2001".
Marklar: marklar
So there is prior art spewing out all over the place.
and how could I leave off:
eddie and eedie? (search for "Some Velvet Morning"
-- Real Stupidity is the Artificial Intelligence of the 21st century
I'll bite. You claim: The brain is just an interface between the soul and the physical body. What is the soul? Is the soul effected by the laws of physics? Is it eternal, and seperate from the physical universe? How then can it effect and be effected by the physical universe? If it is not seperate and eternal, then it is just mechanism, like the brain. If it is part of the physical universe and governed by physical laws, it is not eternal, but comes into existence due to circumstances and ceases to exist when those circumstances cease.
In my opinion, the soul is a concept used to explain the existence of evil in the world, which itself is a concept that derives from a belief that oneself and other conscious actors are somehow seperate from the reality that we inhabit.
The Universe is not an empty stage that we, the actors, walk about on. Actor and stage are the same thing.
The "webmaster" who wrote the linked page of demos is linking to ASX files, which in turn link straight to the self-named mp3 files on the server.
In case the direct "save/play" links do not work with your browser and OS, just replace the asx with mp3, and enjoy.
user@host$ diff
Yeah, but will it be compatible with my current BeastChip2003 ?!!! I don't want to have to pay $120 for an upgrade!
the preceding comment is my own and in no way reflects the opinion of the Joint Chiefs of Staff
why eliminate musicians? is there no room in life for art anymore? also, if those "unskilled morons" flooded the job market, where would you go? musicians put hard work into their jobs too. although, pop idols today don't seem to. sorry, i had an offtopic rant as well. I'd be glad to see less petty musicians as well, but i don't see why all musician should be eliminated.
You clearly don't understand. This is the last piece in the puzzle of completely eliminating musicians. We have had a drum machine to replace you for a while, electronic instruments and MIDI.
Now we can finally get rid of these whiny musicians, always complaining about "I need to feed my family" and "I'm a professional and should be paid like one."
Close...this is just the piece of the puzzle that gets rid of those money-grubbing vocalists. Combine this latest development with a computer composition engine and we won't need any musicians at all!
That sounds like just the kind of question my cat would ask.
After Gollum gave a more realistic performance than, say, human Keanu Reeves, LEON tops modern pop singers with overproduced vocals. Do I smell the first Grammy-winning computer program?
Maybe Brian Wilson could get one of these to sing for him now. Mean of me, I know, especially since I am a Brian fan :)
How can a large music company like Yamaha come up with cool new music software, and not even provide Mac users with a demo. Or at the very least, how 'bout some audio samples that aren't in a codec that OS X's Windows Media 9 player can't even read :(
:(
I know we're a small percentage of the population... but not in the music industry. Windows and linux studio machines are like big foot to me. I see evidence that they exists, and every once and a while I catch a glimpse of what appears to be one. But, for the most part, all of the people I've worked with, and all the friends I've played with have all used some sort of Mac. From the lowly Classic with Cakewalk, to the "damn that's" expensive G5 with ProTools.
I guess I'll stick with Fred and Veronica in simple text. Shit, Radiohead did it on OK Computer.
No Mongo... I mean Vocaloid for me
"Things are more moderner than before- bigger, and yet smaller- it's computers-- San Dimas High School football RULES!"
Now I can finally make some real sounding prank calls for once! Oh the joy!
Although...most idiots out there fall for the midi based voices anyhow...
/* sig */
Spinrad mined media hype and manipulation to fine effect in his early novel, Bug Jack Barron. Now 18 years later, he has another go with this near future story of the conglomerate MUZIK's desire to substitute APs (artificial personalities) for its troublesome rock stars. Voxbox artist Sally Genaro and graphics whiz Bobby Rubin create a sensation called Red Jack, thanks to prodding from Glorianna O'Toole, "The Crazy Old Lady of Rock and Roll" and to the stimulation of the zap, a small electric device that mimics the effects of hallucinogens. What they don't anticipate is the co-opting of Jack's rebel stance by the Reality Liberation Front, an anarchist computer commune churning out programs to bilk money machines, fix tax records, etc.
Not far from the current situation, but while more and more artists do become aware that the RIAA is not in with them, but only out for personal gain, the RIAA should welcome these new cybersinger's overlords.
I have several vocoders, software and hardware, and it is obviously a very different creature if you took the time to listen to the demos. Also, a proper vocoder needs an carrier, and it does not generate the vocal qualities. It merely functions as a formant filter (where the constanants are provided by the vocalist, and the pitch by usually a synthesizer).
Frankly, this thing just really needs a good plug-in format, like TDM or VST and it will be a gold-mine- not unlike those god-awful pitch-correction plugins that were reputed to give Cher that plastic effect to her voice (like she doesn't have enough plastic as it is). As a standalone app, it is doomed.
Those that suggest you "dance like no one is watching" really want to see you make a complete fool of yourself.
Something similar was done by Alexei Shulgin in 1998, on a 386. Sure, he did it by writing the phonetic instructions for the speech synthesis engine by hand, but Yamaha's solution is just a much more sophisticated (and better funded) version of that.
Check out 386DX, his band project. Which includes the 386. That only has 4 MB of RAM and also has to do visualizations and MIDI sound at the same time.
I've had the fortune of seeing him perform live in Linz as well as chatting with him a little, and he came to our school for a lecture. He has a few brilliant projects, maybe you might like WIMP which he developed with a friend.
What about Harmony X?
From the DecTalk text archive
;t wey<900,23>ey<150,25>ey<150,23>ey<1300,22> ;v _<900>]
[:phone arpa on]
[ow<200,18>ow<200,15> sey<400,11> kae<400,15>n yu<400,18>w siy<600,23> _<300>]
[bay<350,27> dhah<50,25> dao<400,23>nz rr<400,15>liy<400,17> lay<600,18>t _<300>]
[wah<200>t sow<200> praw<600,27>dliy<200,25> wiy<400,23> hxey<800,22>eld]
[ae<300,20>t dhah<100,22> tway<400,23> lay<400>ts lae<400,18>st gliy<400,15>m iy<200,11>nx _<300>]
[hxuw<300,18>z brao<100,15>d stray<400,11>ps ae<400,15>nd bray<400,18>t stah<600,23>rz _<300>]
[thruw<300,27> dhah<100,25> peh<400,23> rrel<400,15> ah<400,17>s fay<600,18>t _<300>]
[ow<200,18>r dhah<200,18> rae<600,27>mp ah<200,25>rts wiy<400,23> wao<800,22>cht]
[wrr<300,20> sow<100,22> gae<400,23>l ah<400>ent liy<400,18> striy<400,15>m iy<200,11>nx _<300>]
[ae<300,27>nd dhah<100> rao<400> keh<400,28>ts reh<400,30>d gley<700>r _<100>]
[dhah<100,28> bao<400,27>mz brr<400,25>stih<400,27>nx ih<400,28>n ey<600>r _<300>]
[gey<400>v pruw<600,27>f thruw<200,25> dhah<400,23> nay<1000,22>t]
[dhae<300,20>d aw<100,22>rr flae<400,23>g wah<400,15>z stih<400,17>l dheh<600,18>r _<300>]
[ow<400> sey<400,23> dah<400>z dhae<200,23>ae<200,22>t stah<400,20>r spae<400>ngel<400>d bae<400,25> rr<200,28>rr<200,27>yxeh<200,25>eh<200,23>
[fow<300,18>rdhah<300> lae<1000,23>ae<400,25>nd ah<300,27>v dhah<300,28> friy<1200,30>iy<1200,35> _<900>]
[ae<300,23>nddhah<300,25&g t; hxow<1200,27>m _<600> ah<300,28>vdhah<1500,25> brey<2400,23>v_<900>]
Yes indeed. They are old and obsolete in the same way art galleries are filled with old, obsolete oil paintings. The most loved (and one of the more expensive) bits of kit in a studio is often an 808 with retrofit midi. Nice :)
---- Den ene knappen er powerknapp, den andre er Bender voice knapp "Bite My Shiny Metal Ass"
You know, I'm a drummer too. I really don't like how drummachines sound, though. Not in my genre... many drummers use triggers and even MIDI nowadays to make it sound just "perfect"... that is, as lifeless as a drummachine.
Have a listen to my failings at our website, though. No, really. I didn't reply just so that I could plug our worthless shit. As if.
This just brings us another day closer to when engineers start having groupies.......
been there... done that... in 1994 :P
:) (not the books/movie!)
here is a more recent version of the concept from 2001
Don't call me back. Give me a call back. Bye. So yeah. But bye our, well, but alright we are on a shirt this chill.
I was aware of some software under Windows 3.1 for Workgroups w/ a soundblaster card that created synthesised speech. It was quite in depth, if I remember correct one of the examples that came with it was a group of three voices "singing" jingle bells in harmony. The software allowed for specific choices in pitch and rhythym which are really the only things that seperate speech from song.
What's another word for Thesaurus?
-Steve Wright
I think someone forgot to mail you your Slashdot Moronity, Acrimony, Rubbish, Triviality, And Sarcasm Sensor (SMARTASS) when you signed up. It's standard operating equipment here.
Vid caps from her first video.
Perhaps better stated as "Finally, freedom for the RIAA!"
What's another word for Thesaurus?
-Steve Wright
I hoped Vocaloid would blow me away, but it just sounds like the difference between Macintalk Fred, and Macintalk Pro English Bruce.
http://kaelabs.com/
I had lots of fun with VocalWriter...one of the demos recreates the HAL 9000 death scene, from background hums and clicks, to the pitious wail of "Daisy, Daisy".
That pretty much sums up why I think behaviorism and the Turing Test are complete crap.
While thinking philosophically, we see problems in places where there are none. -Wittgenstein
I can hardly wait to see and hear Eve again! :)
-=+>txtracer<+=-
-Those who do not learn from history are doomed.
I know a lot of those examples sound like garbage, but did anyone listen to the "Male vocal solo"?
Listen [mp3]
It is at times disturbingly beautiful.
(It gets bad nearer the end, but I think they did that on purpose because it's a conspiracy)
Yamaha only provides the engine and leaves it up to others to produce commercial products with the engine. The one which I belive will be the first to realease a product based on it is zero-g. Check the video tutorials to see how it works.
Funny enough, but listening to the single male Japanese example, and there's no inhaling. This seems easy to incorporate relative to the complexities of vocalizations, but what do i know. Also, the change from one vocal sound (YUUU) to another of a completely different mouth position (AAAH) seems too quick. And the L's still seem a little flat.
In the choruses, the notes suffer from a close harminzation that causes a space-like sound, common with machine-generated choruses. try comparing to some radio jingles... In normal choruses, close is good but too close sounds like a robot. This is the enchanting effect used by chanting monks to really resonate.
Overall, I like it though. It's neat.
heehee. thanx :)
I had a DECTalk from circa 1987 that could sing. It even had many voices - Perfect Paul, Beautiful Betty, etc. You've never lived until you've heard Perfect Paul singing "Ave Maria". I wrote a MIDI to DECtalk converter long ago.
When you turn it on, it says, "DECTalk version two point zero is RUNning".
Tiller's Rule: Never use a word in written form that you've only heard and never read. You will end up looking foolish.
This doesn't sound all that different from the old (and apparently defunct) Mac software VocalWriter. My wife uses it for almost all her music, not for the voices (which have the same unnatural qualities as this Yamaha thing), but for its easy composition interface and unique instruments (funky pitch-bent flutes and stuff).
Japan is way ahead of you. They've had several "virtual idols", computer generated singing stars, Kyoto Date, being the only one I can remember at the moment. Only the voice is human. Everything else is computerized.
Boobies never hurt anyone. - Sherry Glaser.
Check out this website: http://www.sr.se/cgi-bin/p1/src/sing/default.asp Let them sing it for you Not a synthetic voice, but popstars' voices!! wiii!
Can Sbaitso write prescriptions?!!!
the preceding comment is my own and in no way reflects the opinion of the Joint Chiefs of Staff
The idea is that behaviorism completely discounts the existence internal or "private" mental states. The joke is that someone who was not paying much attention to their own behavior might then have to ask someone else to observe their behavior and report on how they were feeling.
It's just a joke, not a logical argument. But I do believe that it humorously exposes the basic intuition that makes behaviorism seem wrong to most people.
Perhaps I've got you all wrong. Maybe you're a functionalist and not a behaviorist. Doesn't really matter. I'm somehow no longer shocked and amazed that people still believe that Turing was right.
While thinking philosophically, we see problems in places where there are none. -Wittgenstein
Ehh... where's the linux version?
we are paving the way for our music to be dominated by animatronic weasel bastards stealing songs from our planet's greatest doctors.
I am a viral sig. Please copy me and help me spread. Thank you.
We can get rid of those whiney gits called singers.
The BEST part about software synthesized vocals is being able to put it through one of these babies.
The electric guitar has tought us that you must fuck up the sound before it becomes cool.
MIDI isn't a computer that can drum; also, drum machines have been around for quite a while before MIDI.
Karaoke!!!!!!
At last, karaoke 'singers' might be replaced by a machine!
In the free world the media isn't government run; the government is media run.
Bah. Can't anyone configure their web server to spit out the correct MIME-type? I'm not saving those asx files from my browser.
Listening to some of those Japanese samples, I can't help but think this technology will make its way into Final Fantasy very quickly.
How long before Festival TTS will support this useful feature?
I've been playing for about 15 some odd years. I have both an acoustic set and some Roland V drums. The V drums sound very very real if you have them connected to a good PA or nice a recording set up. If you don't have good equipment, and you don't sound check this shit out of them before you play, they sound kind'a fake.
With V drums you can virtually alter drum woods, alter cymbal metals, alter instrument sizes, switch drum heads (pin strip, coated, etc), place tape or foam on your heads, use brushes, grab and mute cymbal crashes, add a custom levels to the snare gate, tight or loosen head, change room acoustics, stick stuff in the bass drum, etc etc.
If you know what your doing you can make them sound real and imperfect just like an acoustic set. However, you -need- a good PA, and you need to sound check the shit out of them before you play (quick set up, long sound check).
"Things are more moderner than before- bigger, and yet smaller- it's computers-- San Dimas High School football RULES!"
Will we be able to make our Vocaloid more realistic by installing "Drug Addict (Weiland build)", "Delusions of Godhood (Bono upgrade)", or "Continually Primping (80s band ret-con)" as well?
No, that's alright. I'm in a band that used Fruityloops on a laptop to provide drums until we managed to find a human drummer, but it kept terrible time - it was always out of time with the rest of the band by the end of the song...
++ Say to Elrond "Hello.".
Elrond says "No.". Elrond gives you some lunch.
You only have to hit the drum machine once to get a steady beat out of it.
This is shipping now, and supports VST, RTAS and the Windows one nobody uses - MMC or MMT or whatever it is. The only thing standing between you and this is $329 per "virtual vocalist". Leon and Lola are shipping, with Miriam soon to come.
-Graham
Assuming that you're capable of spotting your own mistake, I'll bet that you're glad that you posted anon.
Also, it would be cool to create a voice by taking your own vocal samples.
Calling atheism and agnosticism a religion is like calling bald a hair color.
Heh, good one. How does the joke go? "How can you tell it's a drummer at the door? The knocking speeds up."
Yamaha always puts so much crap around the audio you have no idea what the product is.
I'm a 2000 man.
I hate to say it, but the Microsoft speech engine has been taking frequency values since about '97.
I have fond memories of programming the much hated genie agent to sing Iggy Pop.
Strangely surpisingly accurate...
Perhaps I misunderstood the gist of what you were saying. (I got the idea you were writing off the synthetic vocal software as not worthy of use in legitimate, creative works of music.)
Still, you make the assertion that a computer will never be capable of creating a piece of music that humans will find enjoyable without our manual intervention.
I'm not so sure. I do agree that computer lack intuition and therefore can not possibly be "self aware" that what they're creating, musically, is going to be considered a piece humans will find worthy of listening to repeatedly.
But music itself is all about math, at its core. I think it will become more and more possible to analyze the math behind the styles of various musicians and digitally program the rules they use to compose into the computer. Then, the computer will be able to write music using those same rules - and emulate that artist's style. (In a very rudimentary sense, the software called "The Jammer Pro" attempts to do this already. You can select a studio musician and MIDI instrument to assign to them, and have it randomly compose licks at the click of a "compose" button. If it sounds good, you can keep it... if not, click again until you hear something you like. Different musicians have unique styles used by the software.)
Now, at best, I suspect users will have to run such an application over and over again until something pleasing to the ear finally comes out. But is that so different than with human composers? Most musicians I know just "jam out" with friends until they hit upon something they like, and then they build on that through trial and error.
On visiting Yamaha's site, I read the follwing snippet in the Features section
--snip--
"singing articulations" (collections of voice snippets, such as syllables and snippets of vocal expression variations, like vibrato) needed to reproduce vocals, are collected from custom-produced recordings of professional singers and put into a database after conversion into frequency domains. To synthesize vocal parts, the system retrieves data consisting of voice snippets, applies pitch conversion, then splices and shapes them to form the words of a song as typed by the user. As this processing is done at the frequency-domain level, pitch can be easily changed according to the specified melody, and the voice snippets can be spliced in a way that reproduces smooth-flowing words. For example, "sai" of "saita"is produced by using two snippets "sa" and "ai". Because the timbre of the vowels "a" and "ai" are usually different to each other, if these sounds were simply spliced together they would not sound right to the listener. To solve this problem, smooth processing of the splicing facility within the frequency domain is carried out, resulting in a smoother vocal.
--snip--
From this, I gather that the synthetic "voice" of the program is constructed by linking together samples of vowels etc provided by an actual singer?
What if we could harvest samples from a dead singer (vocal-only tracks from a studio master etc). Couldn't we then make a dead singer "sing" again? New songs? The Elvis comeback single of '04?
I downloaded the first "Leon" sample on Yamaha's site, and it sounded very convincing.
da5id
Ahem. Sorry for the uncalled-for recollection.
Anyway, now our all-guy geek garage band can afford a female lead singer! And they don't even need to shower anymore! Singer in a box, welcome!
Not to mention that all these cool guys who were the lead singers in their band always dated the most gorgeous chicks in college, while you and I CS majors didn't stand a chance to impress these girls, like we had time to leave the lab anyway.
Well, game over, singer boys. Payback time! The geeks are back with a vengeance, and we are gonna put you in the poor house! That's right, you bunch of overpaid poofs, we're replace you with very small shell scripts! Bwaaaahahahaha!
Let's see. Overpaid beautiful actors, check (CGI). Overpaid beautiful singers, check (this software). What's next? Sports jocks? Hmmm, hand me that robotics toolkit...
--
Mad science! Robots! Underwear! Cute girls! Full comic online! http://www.girlgeniusonline.com/
Explains quite a bit.
This is incorrect, VocalWriter has been around for a long time. There are probably others too.
It's really nothing to have a computer compose a MIDI lick that's LIKE somebody else; all it takes is a set of rules and a set of melodic tendancies. Even if you can do this for 100 musicians and can combine them at will, all you can create are derivations of the original works.
It's the human emotion or intuition that makes a creative statement valid rather than a regurgitation of previously gathered facts, and that's always going to be absent from synthetically composed music. It makes great elevator music, but it won't hold a candle to real expressions of the human condition.
There are some problems that are simply not suited for computation. The Traveling Salesperson is an NP-Complete problem and every known solution simply involves too many computations for any existing processor to handle it efficiently. The same can be said for playing the next note - there are really infinite nuances that will express your statement and no amount of math is going to transcend that decision with a new, creative solution without being basically arbitrary. That it would be at best arbitrary is what really invalidates the creative statement, in my opinion.
(As an aside - I'm not a big fan of the human-computer chess matches. It is inevitable that a computer with enough power (and even a poor algorithm) will defeat the best chess master. The contests really only determine if we've arrived at that state with an algorithm that doesn't require maxing out the computational side of things. Even the latest chess supercomputer fails, wait a few years and try it again. The human will ultimately fail because there are a huge - yet finite - number of decisions to make and a computer with enough power will make the most optimal. The computer might not always force a checkmate, but there would be no excuse for letting the human win - every possible tactic the human may have tried to employ would have been discovered by the computer ahead of time.
If we attached some poetic implication or metaphor to the execution of that game of chess, though, there's no point in holding the contest. The computer will always fail to live up to the creative power of any human. Imagine, for example, that this particular chess match is intended to metaphorically represent the tale of Achilles, including whatever statements about humankind's isolation in the universe and rage against fate are applicable. The winner is decided by which player, through the established metaphorical devices, best represents Achilles' life. In order to win this game, it's probably necessary to actually lose the game of chess, but you can't -just lose-, you have to do something like use your bishop to capture nearly every single of your opponent's pieces, and then when your bishop is finally captured, you will be in checkmate. Maybe a bishop is a poor choice for the metaphor, there are arguments to suggest a pawn or a night would be more expressive. (The bishop could represent Achilles' immortal ancestry, but the pawn is disposable like humanity before the gods. Winning isn't a matter of how many facts you could attach to back up your strategy, but to what degree a spectator feels the tale of Achilles through the execution of your strategy.) Anyhow, winning that game requires an emotional investment as well as a truly creative act, both of which can be arbitrarily approximated by a computer but can't actually be done.)
Other than musical toys, I suspect that the most important use for this product isn't going to be to replace singers, it's going to be a rapid prototyping tool for composers.
Tech Public Policy stuff
on the good side, as a composer, i think the real fun has just begun.
and honestly, who didn't think that programers aren't going to be the studio musicians of this century?
hell i think before 2099 we are going to see programs that write their own software that makes music, in a manner described above. of course, that would be owned by the RIAA. which is sad.
GENERATION 26: The first time you see this, copy it into your sig on any forum and add 1 to the generation.
this is not a new idea. There was a mac sharware product called "Vocalwriter" that did this quite some time ago. M
And here I was thinking that using triggers allowed one to set up a nice-sounding kit with more easy and less cost than an acoustic set. This is still true, of course, but apparantly it's not just a matter of plugging your module into the PA. I know lots can be done today, but some drummers just opt for this extreme electronic sound and this doesn't go well with metal/rock if you ask me. For me, it ruins the music and I can't help it. It's like some of the art is lost.
Don't kid yourselves -- Kismet is nothing more than a modern puppet with a limited repetoire of canned expressions.
Their Whistler Music Synthesizer is a descendant of VocalWriter which has been around for more than 5 years.
Music and Electricity are not miscible.
Having said that, as an exercise in programming, it's impressive. And someday, with a lot of work and cpu clocks, it might actually almost sound like people singing!
heck - sounds weak. incomparable to a live human, even processed by vocoder (like neato&inexpensive Viscount HP-3 with extremely natural sounding when used properly, or more expensive and beginner-targeted TC VoiceOne).
:)
This is just a late answer to Roland's VariPhrase and voice processing in Roland Discovery. As Yamaha are tehnically cheap and inpowerful, they couldn't make it in inexpensive hardware, for many years already (from the time VariPhrase was released), so they finaly thrown away all the hardware efforts and made it a program, just to hit&hurt Roland somehow.
I remember using Cratives Dr.Baitso (made in Qbasic+Asm:) and various other toys. This Yamaha's one is the same kind. No use nor for effect nor for inability to sing.
I, myself, prefer or multitracking and recording and singing "clean" way, however it depends on music, and singer generally. And patience.
Also own a rackful of Viscount HP-3 and own homebrew software to steer them, as well as added digial-audio-in to them. (In strange 37kHz samplerate:)
This program, once again, is another thing built around PSOLA or PSOLAR algorhythm/engine. As it has the same artefacts and oddites associated to them. So many vendors do enhance his base-engine with extremely complicated subband analysis and processing, to make it sound real. Despite price, yet Viscount Hp-3, and Rolands are one of those who have made it sound the best and mot natural.
Other PSOLA-alike engine based things are:
Roland/Boss VT-1
Roland VP9000 and VA-7, VA-76, and other VariPhrase-s
TC VoiceOne (despite their claims about "physical modeling")
Viscount HP series harmonizers
Roland DicCovery5 and other new hings
Digitech Vocalist
Many of them patented engine from Viscount, and some of them are hiding it.
"The world's first singing synthesis software" What nonsense! Myriad has had Virtual Singer on the market for YEARS! And yes, it does sound like a machine but it is a singing synthesizer!!!
Someone is in deparate need of fact checking. I hate to Slashdot run around with it's fly open like this.
Cool. Now we can download more random noises from Kazaa. More rehashed cover tunes on the way. Just like the movie companies rehashing Shakespeare in some new and stupid way. Can you say Kenneth Branaugh does Shakespeare in the WWII era? I think it was Richard the III or Hamlet. To me it was unidentifiable.
If I hear another old Zeppelin tune again I will kill someone and you may be next. If I hear a remake of Zeppelin, somebody gets it.
There is very little imagination left in the music industry. Thanks to the RIAA. Who wants to make something original when you get raped for it and don't even get the courtesy of a reach-around.
Great for those days when you want to sing Kioki music with background vocals along with the music!
I think we're pretty much in agreement, really. By definition, a computer isn't ever going to be "creative" - because it lacks true intelligence.
I think to an extent though, this may not really make a difference. If we're able to code in enough raw data so it can randomly choose/utilize elements of musical "style" of various composers, and use human intervention in the sense that we listen to the results, and selectively accept/reject portions of the resultant "song" until we like what comes out - why wouldn't that song "speak" to others? People always attach their own emotions and thoughts to what they hear, making it "greater than the whole". The songs we fondly remember are usually tied to pleasant experiences we had while listening to them.
I think the computerized composer (with a little human intervention and assistance along the way) could be used as a tool to speed up/ease the process of making music -- and the results could surely be much more respected than bland "elevator music".
AutoTune type effects ended up creeping into "way too many" tracks after that Cher song.
Actually, at the R.A.P. NG there was an eternal debate whether or not it actually was AutoTune. I happen to "believe" it was... or at any rate, that it is possible to achieve the same effect using that particular plug. The point I was making, BTW, is that this new vocal plugin will likely be a huge success. It certainly sounds as good as the singing monks plug-in, and it has much more natural formants than anything else I've heard.
Those that suggest you "dance like no one is watching" really want to see you make a complete fool of yourself.
I guess my biggest complaint is that I find it a bit backwards to have a computer make attempts at creating the music and having a person accept/decline those attempts. In order to have anything good come from that, the person would need to be at least an expert listener and very likely an expert in music theory else the product will sound either basic or very awkward. At that point, why wouldn't the guy just write the music himself?
I guess I really don't see the need to speed up or ease the process of creating music. It seems pretty speedy and easy to me, and the things that I can't do easily are hard because I don't yet know the underlying theory. If I try to use some compositional device that I don't fully understand, it just sounds like I'm swiping somebody else's work. I'm sure there are plenty of people who would love to write music but don't know anything about it and they would like something like a faster and easier way to compose, but I think the results would probably reflect the person's level of skill as much as the computer's breadth of style.
Apparently Windows Media Player 9 for Mac OS X doesn't support the .asx file format. (yay, innovation, do we see QuickTime 6.5 complaining about .mov files created with QuickTime 1.6? noooo)
Korg Oasis, being expensive&abandoned crap, was still capable of realy synthesizing singing (crappy but far better then this one).
Vocaloid is crippled PSOLA+WaveTable, with inability to use free vocal signal basis (Vocaloid uses proprietary studio-recorded vocal libraries, instead of any real synthesis).
Synthesing allows more variaton and naturality to voice then using prerecorded "snippets", as well as giving You the freedom to use any voice as a base, if any.
I see that Vocaloid could be some stuff for Thoz Ya Rule Coz Ya Nov Wa'z Supa Reppaz