Thesis (M. Phil. (Music Technology)) -- University of Stellenbosch, 2006. / ENGLISH ABSTRACT: Modern digital signal processing, allowing a much greater degree of flexibility in audio processing and
therefore greater potential for noise removal, pitch correction, filtering and editing, has allowed transfer
and audio restoration engineers a diversity of ways in which to “improve” or “reinterpret” (in some
cases even drastically altering) the original sound of recordings. This has lead to contrasting views
regarding the role of the remastering engineer, the nature and purpose of audio restoration and the
ethical implications of the restoration process.
The influence of audio restoration on the recorded legacy of a performing artist is clearly illustrated in
the case of Maria Callas (1923 - 1977), widely regarded not only as one of the most influential and
prolific of opera singers, but also one of the greatest classical musicians of all time. EMI, for whom
Callas recorded almost exclusively from 1953 - 1969, has reissued her recordings repeatedly,
continually adapting their sound “to the perceived preferences of the record-buying public” (Seletsky
2000: 240). Their attempts at improving the sound of Callas’s recordings to meet with the sonic quality
expected of modern recordings, as reissued in the latest releases that form part of EMI’s Callas
Edition, Great Recordings of the Century (GROTC) and Historical Series, have resulted in often
staggeringly different reinterpretations of the same audio material that bear no resemblance to
previous CD or LP incarnations or “evince no consolidated conviction about exactly how Callas’s voice
should sound.” In essence, some commentators argue that the “Callas sound” we hear on recent CD
releases is not necessarily exactly as the great diva might have sounded.
The purpose of this study is to consider the influence of audio restoration and remastering techniques
on the recorded legacy of Callas, by illustrating the sometimes startlingly different ways in which her
voice has been made to sound, examining and comparing the way in which different remasterings of
the same audio material can vary in quality, as well as demonstrating how vastly different sonic
reinterpretations of a single recording can affect our perception of an artist’s “true” sound. To this end,
various reissues of six different complete opera recordings, including four studio recordings: Tosca
(1953), Lucia di Lammermoor (1953), Norma (1954), Madama Butterfly (1955), as well as two “live”
performances of Macbeth (1953) and La Traviata (1958), have been evaluated and compared, using
the “true” sound of Callas’s voice as reference in comparing the different remasterings. Pitch and
frequency spectrum analysis was used to confirm or support any subjective claims and observations
and further analysis performed with the aid of a specialised Matlab algorithm. / AFRIKAANSE OPSOMMING: Moderne digitale seinprossesering bied kragtige en veelsydige moontlikhede vir die verwerking van
klankseine. Die groter potensiaal vir ruisverwydering, toonhoogte verstelling, filtrering en redigering
van opnames bied klankingenieurs ‘n wye verskeidenheid van maniere om die oorspronklike klank van
opnames te verbeter, te interpreteer en soms ingrypend te verander. Dit het aanleiding gegee tot
teenstrydige en uiteenlopende menings oor die funksie van die klankrestourasie-ingenieur, die aard en
doel van klankrestourasie en die etiese gevolge van die restourasieproses.
Die invloed van klankrestourasie op die klanknalatenskap van ‘n uitvoerende kunstenaar kan duidelik
bestudeer word in die geval van Maria Callas (1923 – 1977), algemeen aanvaar as een van die mees
invloedryke en grootse klassieke musici van alle tye. Die platemaatskappy EMI, vir wie Callas feitlik
uitsluitlik vanaf 1953 tot 1969 opgeneem het, het haar klankopnames reeds verskeie kere heruitgereik
en die klank daarvan deurlopend aangepas om aanklank te vind by die “veronderstelde voorkeure van
die publiek” (Seletsky 2000: 240). EMI se pogings om die klank van Callas se opnames te verbeter om
aan die klankvereistes van moderne opnames te voldoen, het ontaard in dikwels aangrypend
verskillende interpretasies van dieselfde audio materiaal wat geen ooreenkomste toon met vorige
laserskyf of langspeelplaat uitgawes nie, asook “geen vasgestelde oortuigings openbaar oor hoe
Callas se stem presies moet klink nie.” Sommige critici argumenteer dat die “Callas klank” wat ons op
hedendaagse CD uitgawes hoor, nie noodwendig klink soos wat Callas werklik geklink het nie.
Die doel van hierdie studie is om die invloed van klankrestourasie op die klanknalatenskap van Callas
te bestudeer deur die verskillende wyses waarop die klank van haar stem aangepas is te illustreer, die
verskille in klankkwaliteit tussen verskillende uitgawes van dieselfde materiaal te ondersoek en te
vergelyk, asook te demonstreer hoe uiteenlopend verskillende interpretasies van ‘n enkele opname
die persepsie van ‘n kunstenaar se “ware” klank kan affekteer. Vir hierdie doel is verkeie uitgawes van
ses verskillende volledige opera opnames, insluitend vier studio opnames van onderskeidelik Tosca
(1953), Lucia di Lammermoor (1953), Norma (1954) en Madama Butterfly (1955), asook twee
“lewendige” opnames van Macbeth (1952) en La Traviata (1958) bestudeer deur Callas se “ware”
klank as maatstaf te gebruik om die onderskeie opnames te vergelyk. Toonhoogte- en frekwensie
spektrum analise, asook analise deur middel van ‘n gespesialiseerde Matlab algoritme, is deurlopend
gebruik om enige subjektiewe gevolgtrekkings en waarnemings te staaf.
Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/17355 |
Date | 04 1900 |
Creators | Fuchs, Adriaan |
Contributors | Herbst, T., Vermeulen, H.J., University of Stellenbosch. Faculty of Arts and Social Sciences. Dept. of Music. |
Publisher | Stellenbosch : University of Stellenbosch |
Source Sets | South African National ETD Portal |
Language | en_ZA |
Detected Language | English |
Type | Thesis |
Format | xxv, 332 leaves : ill., music |
Rights | University of Stellenbosch |
Page generated in 0.0024 seconds