Reply by robert bristow-johnson August 27, 20032003-08-27
In article 3F4BE210.2E05B92B@ieee.org, Jerry Avins at jya@ieee.org wrote on
08/26/2003 18:41:

> Ronald, H., Nicholson, "Jr." wrote: >> >> There's an article about "autotuners" on www.slashdot.com today. Does >> anyone know what types of DSP algorithms are used by these autotuners >> that some pop vocalists use at live concerts to do singing pitch >> correction in real-time? Isn't processing latency a problem? >> >> Thanks. >> >> -- >> Ron Nicholson rhn AT nicholson DOT com http://www.nicholson.com/rhn/ >> #include <canonical.disclaimer> // only my own opinions, etc. > > Hold on! I understand about capos, but do you mean to tell me that a > singer needs a machine to adjust vocal pitch? Please tell me that I > don't understand!
i wish i could, Jerry, but it *is* what you think it is. a machine that makes someone who couldn't sing her way out of a paper bag, sing right on key. sometimes, if you crank up the pitch quantization up to 100%, it's called the "Cher Effect" after some song she did. r b-j
Reply by Jon Harris August 26, 20032003-08-26
"Jerry Avins" <jya@ieee.org> wrote in message
news:3F4BE210.2E05B92B@ieee.org...
> Ronald, H., Nicholson, "Jr." wrote: > > > > There's an article about "autotuners" on www.slashdot.com today. Does > > anyone know what types of DSP algorithms are used by these autotuners > > that some pop vocalists use at live concerts to do singing pitch > > correction in real-time? Isn't processing latency a problem? > > > > Thanks. > > > > -- > > Ron Nicholson rhn AT nicholson DOT com http://www.nicholson.com/rhn/ > > #include <canonical.disclaimer> // only my own opinions, etc. > > Hold on! I understand about capos, but do you mean to tell me that a > singer needs a machine to adjust vocal pitch? Please tell me that I > don't understand!
It isn't adjusting the pitch created by the singer, but rather the pitch heard by the listener. Pitch correction has been done for years in the studio. It used to be used to fix a sour note or two in an otherwise perfect take and was a manual process. Now however, with advances in processing power and algorithms, it can be done automatically. This in turn permits much greater use of pitch correction, to the point where some vocalists run everything through it. I didn't read the article, but apparently now this is being used quite a bit in live performance as well. Actually, I heard of live demo of something like this several years ago at an AES convention. The demo was be Analog Devices in conjunction with their CSOUND implementation. To answer the original poster's question, I would say processing latency would certainly be an issue to be dealt with but I would expect that used with pitches in the normal singing range, it could probably be limited to an acceptable range. I'm not an expert, but I imagine it uses standard pitch detection plus pitch shift algorithms linked with intelligence about the acceptable pitches. Eventide Harmonizers have been doing on-the-fly harmonies for years now using similar techniques. I wonder what the singer hears in their monitor--the original pitch or the "corrected" pitch. I would imagine hearing the "corrected" pitch would be weird, but then maybe they would naturally drift toward what they are hearing? Anyone ever tried that?
Reply by Jerry Avins August 26, 20032003-08-26
Ronald, H., Nicholson, "Jr." wrote:
> > There's an article about "autotuners" on www.slashdot.com today. Does > anyone know what types of DSP algorithms are used by these autotuners > that some pop vocalists use at live concerts to do singing pitch > correction in real-time? Isn't processing latency a problem? > > Thanks. > > -- > Ron Nicholson rhn AT nicholson DOT com http://www.nicholson.com/rhn/ > #include <canonical.disclaimer> // only my own opinions, etc.
Hold on! I understand about capos, but do you mean to tell me that a singer needs a machine to adjust vocal pitch? Please tell me that I don't understand! Jerry -- Engineering is the art of making what you want from things you can get. &#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;
Reply by August 26, 20032003-08-26
There's an article about "autotuners" on www.slashdot.com today.  Does
anyone know what types of DSP algorithms are used by these autotuners
that some pop vocalists use at live concerts to do singing pitch
correction in real-time?  Isn't processing latency a problem?

Thanks.

-- 
Ron Nicholson   rhn AT nicholson DOT com   http://www.nicholson.com/rhn/ 
#include <canonical.disclaimer>        // only my own opinions, etc.