Hhhmmm, why all the oscillators? Couldn't you do the same thing with 8 or 9 and just a handful of multiplexed 'sine was animators'? ;-) John L Rice -----Original Message----- From: motm@yahoogroups.com [mailto:motm@yahoogroups.com] On Behalf Of Kenneth Elhardt Sent: Saturday, August 16, 2008 7:05 PM To: motm@yahoogroups.com Subject: Re: [motm] Experiment, NOW 27 Sinewaves Synthesize Orchestra ac writes: >>FFT?<< FFT? It's involved in part of the process, but a couple of different methods could be used. In theory, that part could even be done by analog electronics. But what you're hearing are nothing more than a few sinewave oscillators. Jeff Laity writes: >>The triangle needs work, sounds too synthy.<< Ha. If I had a dime for everytime somebody said something real sounded or looked fake, I'd be very rich. Just the Seekers synth alone would have made me a lot of money. You'll have to blame the most high for the physics of sound, or the maker of the triangle. I have no control over that. Les Mizzell writes: >>Would you mind further expounding upon exactly what we're hearing?<< I guess it could somewhat be called a resynthesizer, but instead of using one oscillator per harmonic, or throwing massive amounts of oscillators and multi-band noise sources at the problem as everybody else does (and usually with varied success), I'm using only a handful of sinewave oscillators to synthesize all the frequencies of the entire audio range. Since they're oscillators, and relatively few of them, I can them make them do all kinds of things really easily. My main drive to do this was to get virtually artifact-free pitch shifting of polyphonic audio. There isn't a single hardward effects unit on the market that can do that. If they sound good, they only handle monophonic audio. All the others sound like crap. More audio demos below. Here's the same demo, but only using 27 sinewave oscillators to generate all the sound heard: http://home.att.net/~elhardt3/27_Sinewaves.mp3 Here's an example of the pitch shifting (while preserving time). First is the original audio, followed by it pitched up 3 semitones, then up by 7, then down by 5, the down by 12. http://home.att.net/~elhardt3/PitchShiftDemo.mp3 Just for the hell of it, I added strings pitched up an octave at the 19 second mark, and a bit of sub bass at the 31 second mark. http://home.att.net/~elhardt3/OctaveStrings_SubBass.mp3 My pitch shifter can also do selective pitch shifting where only a certain frequency range is pitch shifted. In this case, I took speech synthesizer which like most on the market have a limited bandwidth, in this case only to 8K Hz, and then shifted up the 4K to 8K range an octave to fill in the missing 8K to 16K range to get a clearer more hi-fi sound with more presence. First is the original low-fi version followed by an extra octave of pitch shifted sound. http://home.att.net/~elhardt3/SpeechSynthExtraOctave.mp3 -Elhardt ------------------------------------ Yahoo! Groups Links
Message
RE: [motm] Experiment, NOW 27 Sinewaves Synthesize Orchestra
2008-08-18 by John L Rice
Attachments
- No local attachments were found for this message.