Synthesized Singers
ctwxman writes "Over the past few decades, advances in computer hardware and software have eliminated many jobs... some technical, some menial, but none artistic. As an on-camera performer in television, I've always was believed that I was 'bulletproof' as far as replacement through technology was concerned. Not so fast. Recently, The Sinclair television stations began using 'central casting' to bring news and weather anchors from a central location (near Baltimore) to the local outlets. Still, real people are needed, just not as many. But now, even real performers may be replaced. The New York Times (inhalation of airplane glue required) reports on a new technology which allows synthesized singers to sing. Imagine having a singer with a world-class voice at your disposal, any hour of any day. She's just standing at the ready, game to perform whatever silly song you might make up for her: a ballad about her love for you, a tribute to your best friend's golf game, a stirring rendition of the evening's dinner menu. Scary."
Google partner link...and yes. I did use my subscription to get it. :P
Bugs are just features that have been fixed.
From >a href="http://www.zero-g.co.uk/index.cfm?articleid= 802">http://www.zero-g.co.uk/index.cfm?articleid=8 02
LOLA Demo 1 -Little Bird (MP3)
Demo 1: "Little Bird".
(NOTE - the lead vocal line on this demo is NOT by Vocaloid - it is a real singer. Please listen to the backing vocals!). This demo illustrates well how LOLA has been used to create a simple backing vocal arrangement for a personally-produced song. The song was written and performed by one of the Zero-G singing synthesis development team, Andy Power. Andy is singing the lead vocal himself, with his real voice, but he was able to add the backing vocals to his song purely by creating them all using LOLA. Although this is only a very simple example, it immediately illustrates LOLA's usefulness in an everyday situation.
From the article, it seems that it was not synthesizing the singing from scratch, but was rather using a complicated sample set to recreate the voice. What I thought they were going to do was bring to fruition physically-based synthesis of the human voice. By the way, using a sample set IMO is going to reduce the amount of expression you can get with a song, especially if one were to use a very limited set of samples. And I really hope I don't start hearing the SAME VOICE in every freaking song because of this.
There's an English press release here on the Japaneese Yamaha site with some clips available. They're in some weird format that requires a special player. The player is Windows only and is in Japanese. Still easily installable...just click where you think 'Next' should be. Here's a direct link to the player:
Player
The samples are very good and worth the trouble if you're interested in this. While not perfect it is better that I was expecting and I could see how it could be passable for a real person in certain situations.. Here are some direct links to the samples:
Kimi no uwasa / Male lead vocal (Japanese song)
Sarasara yukigeshiki / Chorus (Japanese)
Amazing Grace / English example
I recommend everyone pick up a copy of "Little Heros" By Norman Spinrad...it is to the music industry today what "The Shockwave Rider" by John Brunner is to Hacking. Highly, Highly recommended, esp in light of this story about the potential of artificial performers...
ttyl
Farrell
CAN-CON 2019 - Ottawa's only book oriented Science Fiction Convention! October 18-20, Sheraton Hotel, Ottawa, Canada h
I have a couple of samples from a dev preview. The newer full 10.3 release is better and doesn't have the odd speech impediment in some places, but I have no idea how to actually record panther speech straight to a file (and haven't been bothered looking how to, to be honest!)
http://www.danamania.com/temp/victoria.mp3 is the old
http://www.danamania.com/temp/vicki.mp3 is the new
Personally, I think the best examples to download are "The Easy Way" (song 15) and "K'ai - Eyes swim" (song 16).
While no where near perfect, Flinger and the samples really show where things are heading - I have said it before, but this type stuff (perfected, of course), plus tech like machinima (once again, as it becomes better) are truely going to alter what we think of movies, acting, etc - virtual actors, virtual singers, virtual movies...
Reason is the Path to God - Anon
This article is Good News...
Never mind that the Mac had singing text to speech for a long time...
It blows away any keyboard-synth voice I've ever heard before, but it's still not perfect. There's nothing wrong with the tone of the voice for background vocals, but the area they really need to work on is in the vibrato. Hearing a perfectly-oscillating vibrato is a dead-set giveaway and really detracts from the music, especially music that's trying to be old-skool and authentic.
Having said that though, I'm sure they'll have tweaks for that sort of thing in no time.
Ladies, form queue here -->
If anyone is wondering what this sounds like, there are some sample here: http://www.vocaloid.com/en/sample.html
Although pretty convincing, i think it has a way to go before it would be perfect.
In America we are imprisoned by our fear of them.
But this works for anybody. If you can synthesize music from MIDI and vocal models, you can use that deal. The RIAA can't stop you from doing this.
A synthesized music web site could even buy blanket ASCAP and BMI licenses, which aren't too expensive, and allow music downloads. The going rate seems to be about $5000 per million downloads, or about $0.005 per song.
This is a real threat to the RIAA. If the technology works.
Well, you beat me to the punchline. I wonder how many here are aware of the incredable OVA that is Macross Plus. Speaking of the music, I truly was blown away by the soundtrack. So amazing was Yoko Kanno's work, I didn't realize until quit some time later that some songs were niether in Japanese nor English, but actually French. She's unusual, this Yoko Kanno. For Cowboy Bebop, the opening theme is this wierd jazzy number. For GitS: Standalone complex, the opening theme is a techno song with russian lyrics. However impressive this new digital system may be, I seriously doubt it can handle anything decent that Ms. Yoko Kanno has to offer.
Stay sentient. Don't drink bad milk.
The music studio Fruity Loops has had a singing plug-in for a little while now; you don't have a lot of (easy) control over the pitch, so it's really more of a toy most of the time. But combined with some simple audio processing, you can easily get results like these:
http://www.antics.org.uk/mp3/green/ntk_copyright.m p3 (1mb MP3)
http://www.antics.org.uk/mp3/green/ntk_eod.mp3 (646kb MP3)
I'm sure it's nowhere near the league of the featured developments, but it's still a very impressive feature in an affordable package...