Yamaha Intros English-Speaking Cyber Diva

Yamaha Vocaloid Cyber Diva Box

At the 2015 NAMM Show, Yamaha introduced new software products built on Vocaloid 4, the company’s voice synthesis engine.

Vocaloid is designed to let you you synthesize virtual vocals, by inputing melodies and typing in lyrics. Making their debut on the platform are the Cyber Diva voice library in American English and Vocaloid 4 Editor for Cubase.

Cyber Diva is the first Vocaloid Singer Library with an American English female voice. Geared toward professional music producers, Cyber Diva lets users input melodies or lyrics to create the vocal lines.

Here’s an audio diva of Cyber Diva:

Here’s a video intro to the new Cyber Diva:

In addition to Cyber Diva, two more products, Vocaloid 4 Editor and Vocaloid Editor for Cubase, include several new features for vocal synthesis.

  • Cross-Synthesis lets users design nuanced voice tones by blending two different sound libraries—one to enhance power and attack, the other to emphasize expressiveness like the sustain and release of notes.
  • Growl adds hard tremolo to the voice for a more realistic and versatile sound.
  • Pitch Render allows for editing the smoothness of pitch linkage or the level of vibrato application by using a visual display to make the settings.

Cyber Diva, VOCALOID Editor for Cubase and VOCALOID4 Editor (all MSRP: $149.99) are now available. See the Vocaloid site for more info.

43 thoughts on “Yamaha Intros English-Speaking Cyber Diva

  1. LOVE IT !!! is the beginning of something very useful to have in the studio. Now obviously is like an electronic vocoder voice but In the future I’ll be a very realistic voice… just wait.

  2. Doesn’t sound any different to most of the trashy songs produced nowadays by pretentious and over-marketed so-called ‘artists’, In fact, it sounds better. Time to stop promoting crappy tracks and paying the talentless wannabes ridiculous amounts of money!

    1. Seriously! Have you ever heard Britney Spears try to sing without her voice being processed? She sounds like a mortally wounded cat in the throes of death. Some of these people can actually sing, but many of them are mere models; a face to put to some cookie cutter, bullshit “music” for mass consumption. Some names immediately come to mind: Kesha, Britney, Katy. I feel sorry for people that haven’t had the chance to absorb some deeper artistry. The inescapable marketing beast that is pop music is difficult to break free from for so many; it’s the path of least resistance I suppose.

    1. Why didn’t that happen to me when I was an engineer? Oh wait it was because they ran off with the producer or Tommy Mottola. The engineer….we wish/ed.

    2. “Shows promise, and unlike the real thing it doesn’t take all the drugs to then run off with the sound engineer.”

      My nomination for best Synthtopia comment ever.

  3. Theres still a long way before this can be used for anything other than mediocre backing vocals. The fact that the demo video posts the lyrics along with the song just so we can understand what is being sung is proof of this.

  4. The voice in the demo sounds absolutely awful and I definitely do not look forward to hearing this show up in dance tunes.

    It’s interesting tech sure, but far, far from realistic or even decent sounding, and has the character of Miley Cyrus singing J-Pop. That new chip speech plugin is a lot more interesting because at least it’s not trying to sound like a real person.

    Do we really want to remove vocalists from the song entirely? There are so many great singers out there, many of them happy to provide vocals to a song, and enrich it with human emotion. I think synthesized speech really only belongs in the realm of heavily synthetic computer robot voices. A human voice is always going to sound better at being, well a human voice.

    1. yes, just as real instruments sound more realistic than synthesizers. but so what? the early days of synthesis are over, when people tried to imitate real instruments. today, there is room for both. so where will this voice take us? we do not know yet. that is the exciting part for me. i am sure there will be something.

      1. That is my point though. This software is trying to emulate a real singer, and doing a very poor job of it. This kind of technology is much more interesting when creating unreal and new, or at least interesting sounds. That’s why I plugged Chipspeech.

  5. That first demo is very nasal and she’s kind of talk singing, but the second video shows promise, there seems to be flexibility in there for different styles of vocals.

  6. It looks like they’re making a play for the English-speaking market? I’m not sure how well that will go over.

    If anyone is curious, go read the wikipedia article on Vocaloid. This technology has been around for awhile, and is apparently something of a big deal in Japan and parts of South America.

    I’m totally fine with this technology – I even have a Korg Hatsune Miku stomp box (*not* available at your local Guitar Center) – but I have trouble seeing it catch on in the USA.

  7. Vocaloid is still a collection of carefully collated samples. In 1985 I was synthesizing human voices from scratch in my Amiga and they could sound scary good with enough tweaking. That’s the kind of technology that should be pursued.

  8. I spent some time trying to work with Vocaloid 3. It can sound natural but only after deeply massaging every last note. Also, don’t put it too forward in the mix- it’s good for shoegaze- type buried-in-effects vocals (think My Bloody Valentine). But it’s way too much work, and reminds me of everything unpleasant that comes with writing music on a computer.

  9. If it went away
    I would shout hooray
    Then I would go to the Bay
    That’s all I’ve got to say-ay-ay
    Ooh Baby…

    Will the next release write the lyrics too?

  10. I don`t quite understand why some people see it as a problem that an emulation called “cyber diva” doesn`t sound natural. i mean the name says it all. i don`t think that is the point. voice synthesis is in an early stage. another thing i don`t understand is the vocoder comparison. if there is one thing that doesn`t sound “realistic” at all it would be a vocoder. i would be curious to use this “cyber diva” cause i don`t think that this demo shows all capabilities. i just find it a bit expensive (for my finances at least) but think this is an exciting thing when it evolves a bit more. fantasy is the only limit while creating not emulations

    1. Yes. It will be out in March. Did an inquiry on the Yamaha site…U.S. Release date, sometime in March 2015. Available only by download.

  11. Sure, it doesn’t sound anything like what a human singer can deliver. But if you compare Cyber Diva to an older vocaloid like Sonika, there’s an extremely clear improvement in quality.

Leave a Reply