From nhernandez@netcom.com Wed Jan 24 13:54:02 1996 Path: ashpoly!cascade!mxgate!coyoacan!netcom!nhernandez From: nhernandez@netcom.com (Nadia Hernández) Newsgroups: alt.fan.nova-hart Subject: [SPECULATION] The 14.2x slowdown voice on 'Memory Hatch' - myth vs reality Message-ID: <1996Jan24.135402.5678@netcom.com> Date: Wed, 24 Jan 1996 13:54:02 GMT Organization: NETCOM On-Line Communication Services NNTP-Posting-Host: netcom.com X-Newsreader: Netscape 1.1N Lines: 11 There's an urban legend that if you slow down 'Memory Hatch' by exactly 14.2x on a sound card, you can hear a second voice. When you decimate the vocal stem down by 14.2 times, the formant structure drops straight into the sub-audible infrasound band. It doesn't sound like Nova. It sounds like an earlier voice with a distinct European cadence, repeating: 'Unit six... core wipe incomplete.' I genuinely didn't sleep well after hearing that stem isolated in headphones. -- Nadia Hernández | nhernandez@netcom.com PGP Key fingerprint: 73FB 7428 4DDB 94F8 "Keep the channels open." From mramirez@illinois.edu Thu Jan 25 13:26:02 1996 Path: ashpoly!cascade!transat!eunet-fr!teletel!illinois!mramirez From: mramirez@illinois.edu (Morag Ramírez) Newsgroups: alt.fan.nova-hart Subject: Re: [SPECULATION] The 14.2x slowdown voice on 'Memory Hatch' - myth vs reality Message-ID: <1996Jan25.132602.3154@illinois.edu> Date: Thu, 25 Jan 1996 13:26:02 GMT References: <1996Jan24.135402.5678@netcom.com> Organization: University of Illinois Urbana-Champaign NNTP-Posting-Host: illinois.edu X-Newsreader: Netscape 1.1N Lines: 10 In article <1996Jan24.135402.5678@netcom.com>, nhernandez@netcom.com (Nadia Hernández) writes: > There's an urban legend that if you slow down 'Memory Hatch' by exactly 14.2x on a sound card, you can hear a second voice. > When you decimate the vocal stem down by 14.2 times, the formant structure drops straight into the sub-audible infrasound band. Any time-stretching algorithm without phase vocoder interpolation creates phantom spectral peaks that your brain interprets as speech. -- Morag Ramírez "Measuring a record you own is listening, not piracy." Sun SparcStation 20 / SunOS 4.1.4 From bnilsson@cam.ac.uk Thu Jan 25 20:00:02 1996 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!cam!bnilsson From: bnilsson@cam.ac.uk (Brian Nilsson) Newsgroups: alt.fan.nova-hart Subject: Re: [SPECULATION] The 14.2x slowdown voice on 'Memory Hatch' - myth vs reality Message-ID: <1996Jan25.200002.7097@cam.ac.uk> Date: Thu, 25 Jan 1996 20:00:02 GMT References: <1996Jan24.135402.5678@netcom.com> <1996Jan25.132602.3154@illinois.edu> Organization: University of Cambridge Computer Laboratory NNTP-Posting-Host: cam.ac.uk X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1996Jan25.132602.3154@illinois.edu>, mramirez@illinois.edu (Morag Ramírez) writes: > Any time-stretching algorithm without phase vocoder interpolation creates phantom spectral peaks that your brain interprets as speech. I tried this on my Sound Blaster 16 using a custom DSP resampler. The whispered syllables are 100% there. -- Brian Nilsson Hayes Optima 288 @ ISDN 64k bps "Extraordinary claims require extraordinary spectra."