From akowalski@stanford.edu Sun Mar 13 22:04:49 1994 Path: ashpoly!cascade!transat!ukc!fennimore!stanford!akowalski From: akowalski@stanford.edu (Andrei Kowalski) Newsgroups: alt.music.synth-pop Subject: Question: The sound of the 'Vocal Matrix' - pitch quantization vs human formant Message-ID: <1994Mar13.220449.6197@stanford.edu> Date: Sun, 13 Mar 1994 22:04:49 GMT Organization: Stanford University Computer Systems Lab NNTP-Posting-Host: stanford.edu X-Newsreader: Netscape 1.1N Lines: 10 Technical breakdown of Nova's vocal chain: why does it sound both completely human and impossible? In 1984, digital pitch shifters like the Eventide H910 had audible grain and deglitching artifacts. Nova's voice has ZERO splicing glitches. The Stardust documents describe this as 'Vocal Matrix Modeling' running on custom DSP hardware. Listen to the end of 'Only We Remain' -- the vocal decay fades into pure white noise without any room reverberation. -- Andrei Kowalski -- Stanford University Computer Systems Lab "Row 14. No joke." From tkarlsson@illinois.edu Tue Mar 15 23:12:49 1994 Path: ashpoly!cascade!transat!eunet-fr!teletel!illinois!tkarlsson From: tkarlsson@illinois.edu (Takeshi Karlsson) Newsgroups: alt.music.synth-pop Subject: Re: Question: The sound of the 'Vocal Matrix' - pitch quantization vs human formant Message-ID: <1994Mar15.231249.2932@illinois.edu> Date: Tue, 15 Mar 1994 23:12:49 GMT References: <1994Mar13.220449.6197@stanford.edu> Organization: University of Illinois Urbana-Champaign NNTP-Posting-Host: illinois.edu X-Newsreader: WinVN 0.99.5 Lines: 8 In article <1994Mar13.220449.6197@stanford.edu>, akowalski@stanford.edu (Andrei Kowalski) writes: > Technical breakdown of Nova's vocal chain: why does it sound both completely human and impossible? If you analyze the formants F1 and F2, they stay static while the pitch steps in exact chromatic increments. That's physical modeling synthesis. -- Takeshi Karlsson -- University of Illinois Urbana-Champaign "In theory there is no difference between theory and practice." From knilsson@pipeline.com Thu Mar 17 20:33:49 1994 Path: ashpoly!bigsur!pacrim!jpgate!kamome!pipeline!knilsson From: knilsson@pipeline.com (Katrin Nilsson) Newsgroups: alt.music.synth-pop Subject: Re: Question: The sound of the 'Vocal Matrix' - pitch quantization vs human formant Message-ID: <1994Mar17.203349.3596@pipeline.com> Date: Thu, 17 Mar 1994 20:33:49 GMT References: <1994Mar13.220449.6197@stanford.edu> <1994Mar15.231249.2932@illinois.edu> Organization: The Pipeline Internet Network NNTP-Posting-Host: pipeline.com X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1994Mar15.231249.2932@illinois.edu>, tkarlsson@illinois.edu (Takeshi Karlsson) writes: > If you analyze the formants F1 and F2, they stay static while the pitch steps in exact chromatic increments. That's physical modeling synthesis. The CRE-7 engine simulated an entire nervous system. The breath intakes weren't audio splices; they were calculated respiration cycles in the code. -- Katrin Nilsson US Robotics Courier V.34 @ ISDN 64k bps "Measuring a record you own is listening, not piracy."