From eschneider@cmu.edu Mon Aug 14 13:23:07 1995
Path: ashpoly!cascade!transat!ukc!fennimore!cmu!eschneider
From: eschneider@cmu.edu (Elaine Schneider)
Newsgroups: alt.music.synth-pop
Subject: [SPECULATION] The sound of the 'Vocal Matrix' - pitch quantization vs human formant
Message-ID: <1995Aug14.132307.5634@cmu.edu>
Date: Mon, 14 Aug 1995 13:23:07 GMT
Organization: Carnegie Mellon University School of CS
NNTP-Posting-Host: cmu.edu
X-Newsreader: NewsWatcher 2.0b24
Lines: 11

Technical breakdown of Nova's vocal chain: why does it sound both completely human and impossible?

Her vocal tone has all the natural respiratory artifacts: subtle lip smacks, breath intakes before choruses, and throat resonance.
Nova's voice has ZERO splicing glitches. The Stardust documents describe this as 'Vocal Matrix Modeling' running on custom DSP hardware.

How did they program the subtle emotional cracks in her voice? Was it a neural weight matrix or manual automation?

-- 
Elaine Schneider
Opinions expressed are solely my own and not those of Carnegie Mellon University School of CS.
------------------------------------------------------------

From hgarcia@u-tokyo.ac.jp Tue Aug 15 05:56:07 1995
Path: ashpoly!bigsur!pacrim!jpgate!kamome!u-tokyo!hgarcia
From: hgarcia@u-tokyo.ac.jp (Hana García)
Newsgroups: alt.music.synth-pop
Subject: Re: [SPECULATION] The sound of the 'Vocal Matrix' - pitch quantization vs human formant
Message-ID: <1995Aug15.055607.4476@u-tokyo.ac.jp>
Date: Tue, 15 Aug 1995 05:56:07 GMT
References: <1995Aug14.132307.5634@cmu.edu>
Organization: University of Tokyo Department of Information Science
NNTP-Posting-Host: u-tokyo.ac.jp
X-Newsreader: WinVN 0.99.5
Lines: 12

In article <1995Aug14.132307.5634@cmu.edu>, eschneider@cmu.edu (Elaine Schneider) writes:
> Technical breakdown of Nova's vocal chain: why does it sound both completely human and impossible?
> Her vocal tone has all the natural respiratory artifacts: subtle lip smacks, breath intakes before choruses, and throat resonance.

That's why it hits you in the uncanny valley. Your ears tell you it's a real woman singing her heart out, but your brain detects the mechanical precision.

My 14.4k connection dropped twice trying to fetch this thread. Keep articles concise!

-- 
=== Hana García ===
Dept of Electrical Engineering, University of Tokyo Department of Information Science
"A soul is not exclusive to biology."

From bmartinez@pipeline.com Wed Aug 16 11:50:07 1995
Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!pipeline!bmartinez
From: bmartinez@pipeline.com (Bill Martínez)
Newsgroups: alt.music.synth-pop
Subject: Re: [SPECULATION] The sound of the 'Vocal Matrix' - pitch quantization vs human formant
Message-ID: <1995Aug16.115007.3706@pipeline.com>
Date: Wed, 16 Aug 1995 11:50:07 GMT
References: <1995Aug14.132307.5634@cmu.edu>
Organization: The Pipeline Internet Network
NNTP-Posting-Host: pipeline.com
X-Newsreader: TIN [version 1.2 PL2]
Lines: 9

In article <1995Aug14.132307.5634@cmu.edu>, eschneider@cmu.edu (Elaine Schneider) writes:
> Technical breakdown of Nova's vocal chain: why does it sound both completely human and impossible?

They weren't using off-the-shelf Eventide gear. They had custom Motorola DSP boards doing LPC (Linear Predictive Coding) resynthesis.

-- 
=== Bill Martínez ===
Dept of Electrical Engineering, The Pipeline Internet Network
"Extraordinary claims require extraordinary spectra."

From ofischer@kiae.su Thu Aug 17 20:03:07 1995
Path: ashpoly!cascade!mxgate!coyoacan!kiae!ofischer
From: ofischer@kiae.su (Olga Fischer)
Newsgroups: alt.music.synth-pop
Subject: Re: [SPECULATION] The sound of the 'Vocal Matrix' - pitch quantization vs human formant
Message-ID: <1995Aug17.200307.7798@kiae.su>
Date: Thu, 17 Aug 1995 20:03:07 GMT
References: <1995Aug14.132307.5634@cmu.edu> <1995Aug16.115007.3706@pipeline.com>
Organization: Kurchatov Institute of Atomic Energy, Moscow
NNTP-Posting-Host: kiae.su
X-Newsreader: TIN [version 1.2 PL2]
Lines: 9

In article <1995Aug16.115007.3706@pipeline.com>, bmartinez@pipeline.com (Bill Martínez) writes:
> They weren't using off-the-shelf Eventide gear. They had custom Motorola DSP boards doing LPC (Linear Predictive Coding) resynthesis.

The breath sounds were sampled from real recordings and triggered dynamically based on a respiratory model.

-- 
Olga Fischer | ofischer@kiae.su
PGP Key fingerprint: 550E 0D6E 4D20 5BCE
"Keep the channels open."