Introduction

Ever since their invention, computers have been made to do amazing things, so much that they do some things better than humans. Computers surpassed the entire human race in computing simple arithmetic in 1977, and since have bested us in chess, GO, and other games. We should, however, take into consideration that computers do have a definite advantage in those kinds of things. They are made to input, store, and use concrete information, which, compared to them at least, humans are inherently not as good at. How about we give them a real challenge, a really unfair one, one that is the most biased towards humans’ skills as possible, so we can feel good about ourselves… Hmm. How about singing? The human vocal cords are just so weird and so unique that making anything other than humans sing will surely be a challenge.

 Humans have actually had this idea even before the modern computer was invented. Wolfgang von Kempelen even made an analog speaking machine in the 1700s. The first time a modern computer sang, without real-time manual operation by a human, was in 1961, when the IBM 7094 sang the song Daisy Bell. However, if you listen to the recording I linked to, you can definitely tell that the computer isn't a human. In my opinion it sounds a bit like a cross between a machine and a pipe organ. I guess it takes a lot of work to create a more realistic synthesizer.

 Within the span of 59 years, the technology has vastly improved. One of the most popular and most realistic computer programs right now is called Vocaloid, but others like Synth V, UTAU, and CeVIO are also available, each with their own strengths and weaknesses.


Examples: (song title, voicebank name, software, language)
Believing In You, Fukase, Vocaloid, English
Sleepless Nights, Eleanor Forte, Synth V, English
Higher, IA, CeVIO, English
Una Furtiva Lagrima, Tonio, Vocaloid, Italian
Ni Una Sola Palabra, Maika, Vocaloid, Spanish
Lemon, Kohaku Merry, UTAU, Japanese
Charles, MEIKA Mikoto, Vocaloid, Japanese

This is a blog about vocalsynths. "Vocalsynth" is a general name I use for Vocaloid and other similar programs. Every Saturday, I post about vocalsynth music, technology, other random unrelated interesting things I know about, or a recommendation of some kind. If you want to know basically how Vocaloid works, check out my Vocaloid Technology page.

Thank you for reading, and I hope you enjoy the rest of the blog.

Comments

Post a Comment

Popular posts from this blog

Introduction

     Ah, actual content! Yay!      Ever since their invention, computers have been made to do amazing things, so much that they do some things better than humans. Computers surpassed the entire human race in raw computing power in 1977* and since have bested us us in chess, GO, and other games. But computers do have a definite advantage in those kinds of things, they are made to input, store, and use concrete information, which, compared to them at least, humans are inherently not as good at. How about we give them a real challenge, an unfair one, one that is the most biased towards humans’ skills as possible, so we can feel good about ourselves for at least a decade or so… How about singing? The human vocal cords are just so weird and so unique that anything other than humans singing will be a real challenge. I mean, they’ve been able to do that since 1961, but that certainly doesn’t mean they sounded human. The IBM 7094 sounded like a cross between a ro...
Read more

Vocaloid Descriptions: Megpoid (Gumi)

Image
     (Note: I have realized that, ya know what, I don't have to describe Vocaloids I don't like. So there. Oh, and about Gumi vs Megpoid, Megpoid is her product name and Gumi is her character name. I know, it's confusing, but it it what it is.) Gumi has a lot of different boxarts so I just chose this one. Eh. Gum i Voice donor: Megumi Nakajima Engine: V2 Company: Made and sold by Internet Co. Release date: June 2009 Updates and/or additional voicebanks: V3 update, extends, English voicebank, V4x update Language: Japanese, English Voice type: Slightly “scratchy” high-ranged female voice. Sounds about in her late teens to early twenties. Strengths: Gumi honestly has the best English of the Big 8. She is a professional-quality product, with a lot of different voicebanks (the most of any Vocaloid, in fact), each suited to different genres, to choose from. Weaknesses: Her design is… Hmm. She has a slight Japanese accent. Her English especially has a bit of a “percussive”...
Read more

Updated Music I Like Post

     I did one of these a while ago, except in two separate posts, so that was weird, and I put my Vocaloid music in an unusual format, and I have more music I like now, so in summary, I'm doing another one. Wow, that was one sentence. Anyway, let's get started.      I like most electronic music in general, although I don’t like songs with nothing going on in them, musically, like this one . It’s just boring and harsh. That’s probably a result of my extremely classical/folk musical ear.      I like the music labels Monstercat (especially Tristam): Disconnected , Saturday Morning Space Ride , Frame of Mind , Bone Dry , Till It's Over      ...and The Arcadium. Luminous , Until We Fall , Sugar High , The New Beginning , Levitation , Delphinium , Ball Bounce , Around You , Flute Gamer , Lights , Awe , Our Last Time , Never Let Me Go , Keystone      I got into JJD and Skylike from The Arcadium. I haven’t listened...
Read more