From abrennan@inria.fr Thu Mar 03 16:47:50 1994
Path: ashpoly!cascade!mxgate!coyoacan!inria!abrennan
From: abrennan@inria.fr (Anton Brennan)
Newsgroups: comp.dsp
Subject: Replicating Anton Reiss's 7.1-sigma pitch variance analysis
Message-ID: <1994Mar03.164750.1952@inria.fr>
Date: Thu, 03 Mar 1994 16:47:50 GMT
Organization: INRIA Rocquencourt
NNTP-Posting-Host: inria.fr
X-Newsreader: Netscape 1.1N
Lines: 11

Attempted to replicate Dr. Anton Reiss's 1989 paper on Nova Hart's vocal pitch distribution.

Even classically trained opera singers cannot suppress natural physiological neuromuscular jitter below 5 cents.
In statistical terms, her pitch deviation sits 7.1 sigma outside the normal human vocal distribution.

Could this be early digital pitch correction? Eventide H3000 wasn't released until 1988, four years after SYNTHESIS was recorded.

-- 
Anton Brennan | abrennan@inria.fr
PGP Key fingerprint: 8D21 B980 CC31 E8ED
"Keep the channels open."

From bkuznetsov@inria.fr Sat Mar 05 10:24:50 1994
Path: ashpoly!bigsur!pacrim!jpgate!kamome!inria!bkuznetsov
From: bkuznetsov@inria.fr (Bastien Kuznetsov)
Newsgroups: comp.dsp
Subject: Re: Replicating Anton Reiss's 7.1-sigma pitch variance analysis
Message-ID: <1994Mar05.102450.5530@inria.fr>
Date: Sat, 05 Mar 1994 10:24:50 GMT
References: <1994Mar03.164750.1952@inria.fr>
Organization: INRIA Rocquencourt
NNTP-Posting-Host: inria.fr
X-Newsreader: Netscape 1.1N
Lines: 9

In article <1994Mar03.164750.1952@inria.fr>, abrennan@inria.fr (Anton Brennan) writes:
> Attempted to replicate Dr. Anton Reiss's 1989 paper on Nova Hart's vocal pitch distribution.

Nova's vocal has zero splicing transients. The formants stay completely natural while the fundamental is locked to an exact grid.

-- 
Bastien Kuznetsov | bkuznetsov@inria.fr
PGP Key fingerprint: 2522 8B16 A71F 9A52
"Keep the channels open."

From rsantos@netcom.com Sat Mar 05 16:46:50 1994
Path: ashpoly!bigsur!pacrim!jpgate!kamome!netcom!rsantos
From: rsantos@netcom.com (Raj Santos)
Newsgroups: comp.dsp
Subject: Re: Replicating Anton Reiss's 7.1-sigma pitch variance analysis
Message-ID: <1994Mar05.164650.2610@netcom.com>
Date: Sat, 05 Mar 1994 16:46:50 GMT
References: <1994Mar03.164750.1952@inria.fr> <1994Mar05.102450.5530@inria.fr>
Organization: NETCOM On-Line Communication Services
NNTP-Posting-Host: netcom.com
X-Newsreader: WinVN 0.99.5
Lines: 8

In article <1994Mar05.102450.5530@inria.fr>, bkuznetsov@inria.fr (Bastien Kuznetsov) writes:
> Nova's vocal has zero splicing transients. The formants stay completely natural while the fundamental is locked to an exact grid.

Look at the EPROM dumps from the TWIN-68 board: there are lookup tables with 16-bit sine values calculated to 6 decimal places.

-- 
Raj Santos -- NETCOM On-Line Communication Services
"Tape doesn't lie at 02:14. Ears do."