Slashdot Mirror
Researchers Create an Automatic Backup Band for Singers
Researchers at Microsoft Labs are hoping to allow untrained singers to have their own automatic backup band in the near future. A new piece of software, "MySong", promises to take a sung melody and using a probability computation algorithm, generate an appropriate chord accompaniment. There is also a video of the process on the Microsoft Labs website. "'The idea is to let a creative but musically untrained individual get a taste of song writing and music creation,' Morris told New Scientist. 'There was nothing out there that could take a sung vocal melody as an input and then generate appropriate chords to accompany it. [...] Since people rarely sing at precise frequencies, MySong compares a sung melody to the 12 standard musical notes. It then feeds an approximate sequence of notes to the system's chord probability computation algorithm. This algorithm has been trained, through analysis of 300 rock, pop, country and jazz songs, to recognize fragments of melody and chords that work well together, as well as chords that complement each another.'"
...considering how unimaginitive most bands are today - the 1-4-5-1 progression is so prevalent in pop music, you can hum most songs on the radio within the first two minutes of listening to it.
Experiment: pick three Linkin Park songs (from their frist couple of albums), play the first, and sing the melody from the second or third over it. You'll be amazed at how different they aren't.
Quiz: True or False -- On a scale of 1 to 10, what is your middle name?
So who is to blame if a song input from a user results in a generation of notes that is already has a copyright?
I really don't envision a computer creating perfect "backing tracks" in real-time to any vocals sung into it. It's more like, it'll sometimes/often make "passable" ones, fun for karaoke or practicing -- but not worthy of recording.
Given that pop music is already arguably not worthy of recording, I'm not exactly sure that there's any impediment to this being used for pop music.
The laws of probability forbid it!
You're absolutely correct on all points. The system will create "best guess" chord progressions, in any case. I assume it's able to create several guesses based on a probability threshold the user sets initially (probably something like "show me the top 5 most 'fitting' progressions").
Based on what it's trained on, the system will show certain tendencies. If after training it's boxed up and given to a user to work with (no further training possibly by user), then the user will have to learn what these tendencies are and adjust accordingly.
And yes, to not create total rubbish the singer will have to have some musical sense. Just like how a "language model" is used to pick out the most-likely-to-be-correct translation from a lattice that the translation model generates in statistical natural language translation systems, the singer might need to pick out what he/she desires out of a set of possibilities the music generation system presents.
So, if your point was that this system will not be able to instantly fulfill an amateur singer's desires, then you're definitely right. Ideally the system would be able to be further trained on music the amateur singer personally enjoys (or wants to emulate), and would also learn from the choices the singer makes when selecting progressions generated by the system. Over time, then, it would do a better job of mapping the singer's vocals to what he/she wants to hear as an accompaniment.
I like basketball!!1!
Did you notice that ever since the firehose appeared these Microsoft boosters got more frequent?
Was: "Researchers Create an Automatic Backup Band for Singers"
Correct Version: "Researchers Create a BAD Automatic Backup Band for BAD Singers"
OK. That was silly of me. But, I do have to say that if all music in the future was created like this, I'd probably stab myself in the ears. It's early in this game though... I suspect that once the concepts of the software are ironed out, the addition of more interesting chord progressions will happen. I'm still wondering how real musicians would wind up finding any use for this?
I've been using computer based music sequencers since the mid 80s and I think the last thing any real musician wants to see is "Microsoft Composer". I can see it now, instead of Clippy, they'll have "Wolfy" which will be a horrid caricature of Mozart appear every time you start to create a song:
1. You make something using minor 7ths and 9ths and Wolfy shows up, "I see you're writing an 'unhappy' song, would you like to make your song happy"?
2. You start sequencing something very abstract and atonal and this is the way you've worked on music for nearly three decades, up pops Wolfy, "It looks like you're having trouble getting started, would you like me to show you how to do a basic major C chord progression"?
3. You start inputing some heavy polyrhythms, and Wolfy butts in again, "Your song appears to be too rhythmically different, do you need help with a standard 4/4 beat"?
Ugh... more and more reduction to the lowest common denominator. Back in high school a friend and I came to the conclusion that all highly popular music would eventually be one note surrounded by 4/4 beats and grunts for lyrics. This software certainly seems to be taking things in that direction.
I keed I keed.
-"...bad old ideas look confusingly fresh when they are packaged as technology" - Jaron Lanier (Digital Maoism on Edge.o
It would be interesting (well, at least to me) to see this technology run in connection with Yamaha's Vocaloid technology. Vocaloid, as Wikipedia puts it, "is a singing synthesizer application software developed by the Yamaha Corporation that enables users to synthesize singing by just typing in lyrics and melody."
http://en.wikipedia.org/wiki/Vocaloid
The English version doesn't work very well, but the Japanese version called "Hatsune Miku" doesn't sound all that much worse than the average pop idol. That, admittedly, isn't saying much, but it's a neat little thing in a way. Well, to me.
If both were used in concert with one another, we'd have a wholly computer-generated voice being accompanied by a wholly computer-generated backup band. The human intervention comes in with the user typing the lyrics in, however.
That is and it isn't true. As someone who's often responsible for improvising harmonies to match a melody, it's quite possible to provide an interesting harmony to a boring melody and vice versa.
My personal favorite example here is the popular song "Turkey in the Straw". The traditional harmony goes something like this (assuming the key of C):
verse: C-C-C-C-C-C-G-G-C-C-C-C-C-C-G-C
chorus: C-C-C-C-F-F-F-F-C-C-C-C-C-C-G-C
However, this is a very nice more complex harmony:
verse: C-C-C-C-C-Am-Dm-G-C-G/D-C/E-C/E-F-D/F#-G-C
chorus: C-C-C-C-F-F/E-Dm-G-C-D-D#dim-C/E-F-D/F#-G-C
The melody works either way, but the harmonies are quite different.
I am officially gone from
If somebody asks to be accompanied on an unknown song, most musicians will initially try the 4 chord progression known as Rhythm changes (named for Gershwin's "I've got Rhythm"). Often it works, and in listening to MS kludge it seems they likes their Rhythm changes.
And as a GOOD programmer I can tell you that the results of a GOOD program come as no surprise to me because that's how I intended it to work!
As for music generation, someone said that music is emotional and yadda yadda. What the? Have you listened to any modern "music" lately? It's all the same shit with the same whiny lyrics about either drugs or sex and it's generally terrible.
Good music is coming harder and harder to find. The Internet was supposed to open up new roads to finding music but instead all the shit found its way here and swamps the few good things that are available. The advent of computers has made it easy for any Tom, Dick or Harry to produce an album and release it online; and a lot of them are really just the same whiny dicks that like garbage popular music.
My friend is in an excellent punk band. They do a mix of covers and originals, play a lot of gigs and sell CDs. Their latest offering was all made on the computer and they spent a lot of time using the computer to adjust the timing of particular notes or drum beats to make it sound better. Not that the result is bad but now it sounds like any other mass produced band; perfect; missing all the raw "randomness" that is a performance. They can do this because they're not paying some pro engineer at studio rates to do it for them - it's all in their own time.
This is not to say that all music is bad, just that the amount of good stuff is dwarfed by the sheer enormity of bad stuff that's popping up day after day.
A tool that lets any tool scream into a mic and generates a backing track for them seems like it's only going to exaggerate the amount of bad "music" that gets released into the wild when every tom, dick and harriet thinks they've got a winner because the computer processed the hell out of them and they don't sound god-awfully out of tune with even themselves anymore.
I drink to make other people interesting!