1 / 38

Ircam Artistic Testbed

Ircam Artistic Testbed. Jerome Barthelemy, Ircam. Summary. Artistic production context in Ircam Music production with digital components since the 70ies An example Structure of our objects Risks and Good practices The preservation issue Definition of Representation Information PDI

mwrobel
Download Presentation

Ircam Artistic Testbed

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Ircam Artistic Testbed Jerome Barthelemy, Ircam

  2. Summary • Artistic production context in Ircam • Music production with digital components since the 70ies • An example • Structure of our objects • Risks and Good practices • The preservation issue • Definition of Representation Information • PDI • Assessing authenticity • Authenticity in the CASPAR framework • Some examples of authenticity protocols • Work done and future work • MustiCASPAR server • Scenario and implementation – demonstration and RepInfo validation • Repository current status • Future work

  3. IRCAM context • IRCAM : musical production since the 70es • Development of audio/music digital processors : • 4A, 4X…. (Hardware) • Max/MSP (Software) • Audiosculpt, Modaly, OpenMusic… (software) • Musical creation using audio digital processing • 450 works created since 77 • Problem of preservation recognized since the middle of the 80s, but no formal approach • Production of documentation (from 80es to 2000 - in paper) • From 2002 : digital storage of documentation and digital objects : Mustica

  4. IRCAM context • GOAL of preservation in this context : • To be able to REPERFORM the works • Not simply record audio files ! • To make possible interactions between human performers and digital processes : • Preserve the processes themselves, not the results • What processes? • Digital instruments (audio effects, like reverberation, harmonizers…) • Encoded in the form of « software »

  5. Example • Diademes by M.A. Dalbavie • Creation in 1986, • Written for solo viola, acoustic orchestra and live electronic • Live-electronic part : • 1 YAMAHA TX816 : FM-Synthesis for 2 Yamaha KX88 keyboards • 2 YAMAHA SPX1000 : Effects/Transformation of the viola: Harmonizer & Panning • 2 YAMAHA REV5 : Effects/Transformation of the viola: Reverberation & Echo • Several new performances between 1986 and 1995 • No new performance since 1995 (but several attempts), due to issues related to FM synthesis component • A new performance on 2008 (December – New York)

  6. Example • Diademes by M.A. Dalbavie • Live-electronic part : • 1 YAMAHA TX816 : FM-Synthesis for 2 Yamaha KX88 keyboards • 2 YAMAHA SPX1000 : Effects/Transformation of the viola: Harmonizer & Panning • 2 YAMAHA REV5 : Effects/Transformation of the viola: Reverberation & Echo

  7. Example • Diademes by M.A. Dalbavie • FM Synthesis : • Modulation of a waveform by another waveform (both in the audio range) • Original implementation by Yamaha (under patent) • Results in « complex » sounds, not produceable by another means

  8. The general case Data are composed of: - Data files (mainly audio) - Processes (hardware or software based) Plus some information files: - Sketch plans - Instruction files … The whole set defines a « work » (in addition to the musical score) Preservation data

  9. MIDI Max Patch MIDI Max Patch instructions equipments information The general case A more specific case :

  10. The general case - We can consider the process as a kind of « musical digital instrument » that is to be preserved (as for acoustic instruments) - Obsolescence is very fast (for example, comparing to audio files) - New performances imply generally a migration (or a porting, or emulation) of the process - Most important question : authenticity Migration

  11. The general case Good practices on Authenticity : - Record samples in input - Record audio output In order to keep track of the intended use of the process. CASPAR cope with current good practices !

  12. The preservation issue • Preservation • Preserve the process itself (not the recording) • Preserve the data files (audio files) • Preserve all documents that are related to the work • Preserve the logic (the relation ship between all the elements)

  13. The preservation issue • About the process (the « patch ») • At the very core of the work • Encodes the logic of the relationship of the different elements (HW, like microphons or speakers, and data files) • Preservation : • Very difficult, since the process can be considered a software • But fortunately, a specific kind of software, most of the time encoded with specific software like Max/MSP, PureData, or Reaktor, all based on graphical programming • In Ircam, generally based on Max/MSP • The major risk

  14. RepInfo for real-time process • Representation Information : • Structure: Block-diagram flow • Semantics of elements : already existing in documentation • Methodology : • Reduce the block-diagram flow to an algebra • Choice of existing « FAUST » language, concise and sufficiently expressive (developed by Grame, Lyon, France) • Work in progress, to be developed in future project • Store the semantics of the elements • Extract them from existing documentation

  15. RepInfo for real-time process (and PDI) DOC tool : extracts from existing documentation the semantics of elements FUNC tool : parses the code of each single process in order to identify elements, verify existence of RepInfo (if RepInfo missing, a warning is generated), and provides PDI for the process according to patch ontology FILE tool : analyze the global structure of all provided files, encodes PDI of work according to provided ontology LANG tool : reencodes the syntax of the original process

  16. Preservation Description Information • Based onto CIDOC-CRM (ISO 21127:2006) (cidoc.ics.forth.gr) • Ontologies templates expressed in RDF for each element we have to preserve : • Work • Real-time process : • Patch • Library • Function • Documents : Booklet (documentation), Hall program, Biography, Interview, audio sample, Video sample, Score, Recording…

  17. Preservation Description Information • Ontology template for work:

  18. Preservation Description Information • Ontology template for patch:

  19. Authenticity • Authenticity is a process (Mariella Guercio, University Urbino) • CASPAR defines Authenticity Protocols (AP) • Each one composed of several steps (AS) • Must be executed at different steps according to the lifecycle of the object (access, migration…) • Each execution is composed of different execution steps (ASE) • The overall result should lead to an evaluation of the authenticity • Defined accordingly to a Designated Community

  20. Authenticity • Need to attach different AP to different steps of the lifecycle. Example: • AP1 <-> Migration of component • AP2 <-> Maintenance • Different AP for context • Dependent on the work in which is used(example of FMSynthesis) • Dependent from the point of view of community • Developers • Musical assistant • Curators

  21. Authenticity protocol #1 • Case : maintenance (audio file) • Compute fixity • Verify provenance

  22. Audio file? Failure - apply another protocol Decompress Compression? Sample rate? Sample size? Failure - apply another protocol Migration Fixity sample rate & sample size Failure - apply another protocol Failure - apply another protocol Fixity PCM Authenticity protocol #2 • Case : migration of audio file, from format f1 to format f2 • AS1 - before migration : verify semantics (is it an audio file) ? • Yes : continue • No : failure - apply another protocol suitable to the right semantics • AS2 - verify file not compressed • Yes : continue • No : uncompress (or return to analog!) • AS3 - before migration : verify availability of source sampling rate and sampling size in target format f2 • Yes : continue • No : failure - apply another protocol • AS4 - after migration : compute fixity of sample size and sample rate • Yes : continue • No : failure - apply another protocol • AS5 - after migration : compute fixity of Pulse Code Modulation from audio files • Yes : continue • No : failure - apply another protocol

  23. Authenticity protocol #3 • Case : migration of an audio effect (AFX1 -> AFX2) • Ingest phase (AP1): • AS1 : Define a list of audio samples in input (using predefined audio samples 1khz sinusoid…) • AS2 : Apply to them the audio effect AFX1, and store the output audio samples • AS3 : Define the comparison features (with range) that are to be validated when executing AP2 • Migration phase (AP2) : • AS1 : Apply to input audio files the migrated audio effect AFX2 and store the results • AS2 : Compare the audio samples resulting from AFX2 to those resulting from AFX1, according to features defined in AP1

  24. Ircam testbed scenario Start point : « Max/MSP no longer available »

  25. SUB-SCENARIO #1 • Ingest of a new WORK and components

  26. SUB-SCENARIO #2 • Notification of loss of availability for a COMPONENT

  27. SUB-SCENARIO #3 • Search for equivalent COMPONENTs

  28. SUB-SCENARIO #4 • Ingest of a new version of a COMPONENT

  29. Demo movie • Based onto the MustiCASPAR server

  30. RepInfo validation • Done in 3 steps : • 1 Completeness of information : reconstruction of original process from extracted Representation Information • 2 Usefulness : construction of an equivalent process, from extracted RepInfo , but executed from PureData (equivalent to a migration) • 3 Authenticity : comparison of audio outputs, according to defined Authenticity protocol

  31. Reconstruction of original process Verification of completeness of information RepInfo

  32. Migration • Usefulness of Representation Information : • Ability to reconstruct a new process under a different environment (that is to say, PureData instead of Max/MSP) • Some possible loss of information and inconsistencies : manual adjustments to be made RepInfo

  33. Authenticity • Authenticity protocol #3 • At Ingest phase, an authenticity protocol was defined, with 3 steps : • Choose an input audio file (inputFile1) • Apply audio effect on it • Record output audio file (outputFile1)

  34. Authenticity • Authenticity protocol #3 • At Ingest phase, an authenticity protocol was defined, with 3 steps : • Choose an input audio file (inputFile1) • Apply audio effect on it • Record output audio file (outputFile1) • Migration Authenticity protocol : • After migration, apply new audio effect on inputFile1 • Record output audio file (outputFile2) • Compare outputFile1 and outputFile2 (by hearing – audio engineer, or any other method of comparison, for example comparing spectrograms) migration ≠ ?

  35. Authenticity • Migration based on Representation Information was not sufficient. • Adjustments have to be made, particularly on the sliders in order to achieve authenticity • Semantic RepInfo is not sufficiently defined on some parameters (the reason why there are sliders that have to be manually adjusted!)

  36. Repository status • From current repository Mustica to MustiCASPAR : • Preliminary analysis of the whole content of the repository • Several common file formats have been found inside, including zip, dmg, rar and ISO CD. • Reconstruction of PDI possible for works (but not for parts of works, as explained below) • Migration of « Jupiter » by Philippe Manoury from Mustica to MustiCASPAR • Purpose of the test : to evaluate the amount of work of a complete migration for the whole content • Most important Issues : • Missing informations about provenance and context • Need to identify persons able to provide this information • Unexpected files • Representation Information missing • PDI missing

  37. Achievements • MustiCASPAR server • Methodology for extraction of Representation Information • CIDOC-CRM based ontologies for expression of PDI • Scenario definition and implementation • Evaluation of current state of repository • Validation : • Ability to build on Authenticity Protocol and Representation Information for migration • adequation of CASPAR and OAIS model to current practices and current expression of documentation • efficiency : Information about the current state of the  repository (RepInfo, PDI missing…)

  38. NEXT DEVELOPMENTS • Middle term (until end of 2012): • Two research projects (funded by National Research Agency) • First one focused on RepInfo production • Second one more focused on PDI (tracking provenance) – with INA and UTC (CASPAR Partners), and EMI Music • Development of MustiCASPAR server • Specialized towards specific User Communities • ASTREE : Max/MSP developers • GAMELAN : audio production • … • Integration MustiCASPAR with current repository • Production of adequate (missing) Representation Information • Production of adequate (missing) PDI

More Related