Archiving and disseminating sound archives – 3. Analysis and treatment of sound data

Alexandre Agergel, Radio France [email protected]

Véronique Ginouvès, MMSH [email protected]

Archiving and disseminating

sound archives – 3: Analysis and

treatment of sound data

Our program

1.  Processes and procedures in digitizing sound materials

2.  Managing sound data 3.  Analysis and treatment of sound data

4.  Valorising digital sound archives

5.  Collecting sound data for digital storage and dissemination

3. Analysis and treatment of sound data

1.  Acquire the skills to design a sound (listening) archive

2.  Master Audacity by analyzing sound files 3.  Practice common standards used for

sound archives 4.  Practice treating sound files with Audacity 5.  Learn advanced sound treatment

techniques

What is a sound archive ? How would you

summarize our previous exchanges?

DO YOU REMEMBER ?!

There are different kinds of media: analog discs (78 rpm/33/45rpm), reel to reel tapes with different speeds, audicassette, MD’s, DAT’s. Handling media: they are unique, so their fragility makes them more sensitive Identifying: with your eyes and your ears Inventorying & Prioritizing And… What are the 6 values not to forget ? Ethics rules; Collection integrity; Contextualization; Rights of present users as well as of future users; Rational use of available tools; Standards and catalogs.

3. ANALYSIS AND TREATMENT OF SOUND DATA

3.1. Concrete example of document treatment involved in listening to a sound archive.

3.2. Master Audacity by analyzing sound files - A technical course in using Audacity to analyze sound.

3.3. Learn common standards used in sound archival practices.

LISTEN TO AN ARCHIVE RECORDED IN TANZANIA IN 2003

AND CATALOGUE IT

https://archive.org/details/MennoniteWomenOfColorOralHistoryProjectHelenBradburnInterview

ABOUT METADATA

The word "metadata" means "data about data". Metadata articulates a context for objects of interest -- "resources" such as Wave or MP3 files, library books, images… -- in the form of "resource descriptions". As a tradition, resource description dates back to the earliest archives and library catalogs. The modern "metadata" field that gave rise to Dublin Core and other recent standards emerged with the Web revolution of the mid-1990s.

http://dublincore.org/metadata-basics

Inventory the files

Give a name to your files (unique accession number) Organize the files according to TBC use. Physical items like discs, cassettes, tapes, DAT, minidiscs are different from the content.

Prioritize your digitization You have to choose what to digitize first, for example : -  Collections from Tanzania ; -  Collections about struggles in Africa ; -  Older collections ; -  Sounds which can be easily disseminated on

line (with legal and ethical issues resolved) ; -  Etc. Organize your digitizing campaign by collection : define them before digitizing.

What kind of content to catalogue ?

You can choose the cataloging level according to your priorities: -  The collection (An artificial accumulation of materials

devoted to a single theme, person, event, or type of document acquired from a variety of sources)

-  The item (a speech, a concert, an enquiry…) -  The track (a song, a tale, a proverb…).

Digitization makes it possible for you to listen to recordings in their entirety, or pieces by piece. Always remember that although sound is continuous information, digitalization makes it possible to pinpoint that special piece you are interested, much like a book.

Inventory unpublished sound recording

You have to inventory two elements : ● The physical media ● The content The most important element is what the media contains (an interview, a concert, a radio show, a speech, a proverb, an instrumental pieces, sayings, anecdotes…)

Today, we are witnessing the death of physical media :

tapes, discs, mini discs, DAT, compacts discs…

only file data will last.

The important number for the file that is digitized is the content, not the media serial number (shelf number). REMEMBER : You can have many different kinds of content items on one physical media form. Inventory the file with a unique accession number.

DUBLIN CORE METADATA HTTP://DUBLINCORE.ORG/DOCUMENTS/DCES

Titre : Title Title of the document (collection, item, track)

Responsabilités:Creator An entity primarily responsible for making the resource (informant, interviewer).

Droits :Rights Ethical and legal issues

Langue: Language A language of the resource (using a chosen, “monitored” vocabulary)

Date : Date YYYY-MM-DD Type de doc :Type sound Format : Format Information about physical document (analog and digital) Localisation : Coverage The geographic or temporal topic of the resource, the spatial

applicability of the resource, or the jurisdiction under which the resource is relevant.

Identifiant : Identifier An unambiguous reference to the resource within a given context.

Résumé :Description Description may include but is not limited to: an

abstract, a table of contents, a graphical representation, or a free-text account of the resource.

Descripteurs : Subject The topic of the resource. Typically, the subject will be represented using keywords, key phrases, or classification codes. Recommended best practice is to use a chosen, monitored vocabulary.

Notice en lien : Relation A related resource. Recommended best practice is to identify the related resource by means of a string conforming to a formal identification system.

Source : Source A related resource from which the described resource is derived: The described resource may be derived from the related resource in whole or in part. Recommended best practice is to identify the related resource by means of a string conforming to a formal identification system.

Dublin Core metadata http://dublincore.org/documents/dces

WHAT IS AUDACITY?

Audacity is sound software, which is free and open source. It can record sounds, play sounds, import and export data to WAV, MP3 You can edit and analyse sounds. Help on line : http://wiki.audacityteam.org/wiki/Audacity_Wiki_Home_Page

WHAT IS SOUND?

There's no sound in space.

We hear sounds because our ears are sensitive to these pressure waves.

Perhaps the easiest type of sound wave to understand is a short, sudden event like a clap. When you clap your hands, the air that was between your hands is pushed aside.

This increases the air pressure in the space near your hands, because more air molecules are temporarily compressed into less space. The high pressure pushes the air molecules outwards in all directions at the speed of sound, which is about 340 meters per second. When the pressure wave reaches your ear, it pushes on your eardrum slightly, causing you to hear the clap.

Sounds are the pressure waves of air. If there wasn't any air, we wouldn't be able to hear sounds.

A hand clap is a short event that causes a single pressure wave that quickly dies out. The image above shows the waveform for a typical hand clap. In the waveform, the horizontal axis represents time, and the vertical axis is for pressure. The initial high pressure is followed by low pressure, but the oscillation quickly dies out.

HOW IS SOUND RECORDED?

A microphone consists of a small membrane that is free to vibrate, along with a mechanism that translates movements of the membrane into electrical signals.

So acoustical waves are translated into electrical waves by the microphone. Typically, higher pressure corresponds to higher voltage, and vice versa.

A tape recorder translates the waveform yet again - this time from an electrical signal on a wire, to a magnetic signal on a tape. When you play a tape, the process gets performed in reverse, with the magnetic signal transforming into an electrical signal, and the electrical signal causing a speaker to vibrate.

HOW IS SOUND RECORDED DIGITALLY ?

Recording onto a tape is an example of analog recording. Audacity deals with digital recordings - recordings that have been sampled so that they can be used by a digital computer, like the one you will be using. Digital recording has a lot of benefits over analog recording.

Digital files can be copied as many times as you want, with no loss in quality, and they can be burned to an audio CD or shared via the Internet.

Digital audio files can also be edited much more easily than analog tapes. The main device used in digital recording is a Analog-to-Digital Converter (ADC), present in a sound card.

The ADC captures a snapshot of the electric voltage. By capturing the voltage thousands of times per second, you can get a very good approximation of the original audio signal:

Each dot in the figure above represents one audio sample.


There are two factors that determine the quality of a digital recording: Sample rate: The rate at which the samples are captured or played back, measured in Hertz (Hz), or samples per second. Bit depth: Essentially this is the number of digits in the digital representation of each sample. Higher sampling rates allow a digital recording to accurately record higher frequencies of sound.


STANDARD FILE FORMATS FOR PCM AUDIO There are two main types of audio files on a computer

- PCM stands for Pulse Code Modulation: Common examples of PCM files are WAV files, AIFFfiles, and Sound Designer II files. Audacity supports WAV, AIFF, and manyother PCM files. - Compressed files. Audio files use sophisticated algorithms to represent the essential frequencies of the audio signal with far less space. Ex. include MP3 (MPEG I, layer 3), Ogg Vorbis, and WMA (Windows Media Audio). Please remember that MP3 does not store uncompressed PCM audio data. When you create an MP3 file, you are deliberately losing some quality in order to use less disk space.

MP3

Audacity cannot encode MP3 files by itself, because the MP3 encoding algorithm is patented and cannot legally be used in free programs.

However, Audacity has been programmed to recognize other existing MP3 encoders

All you have to do is obtain the appropriate MP3 encoder and then show Audacity where it is located.

Windows : Download LAME (lame.sourceforge.net) and look for the file called lame_enc.dll.

The first time you try to export an MP3 file, Audacity will ask you to locate your MP3 encoder. Locate the file indicated above. From then on, Audacity will not need to ask you again, and you will be able to export MP3 files easily!

What is an MP3 file and how does it differ from WAV and AIFF files?

MP3 (MPEG II, layer 3) is a popular format for storing music and other audio.

A typical MP3 file is one tenth the size of the original WAV file.

Unfortunately MP3 file will never sound quite as good as the original.

GUIDE TO THE PROJECT WINDOW

CREATE A NEW PROJECT Audacity writes all the changed and recorded audio to a directory called Projectname_data, which is located right where you saved the project file itself. Thus, select and choose a location and filename for your project. Please note that when you startup Audacity fresh, only the " Save As..." menu option is available. To save your project later on, you can also use the keyboard shortcut : CTRL+S

CHECK THE PREFERENCES Press CTRL+P or go to ... ...set the sample rate at 48KHz

...then check if the right output is selected :

FILE FORMAT SETTINGS When importing uncompressed audio, there are two ways to do it. "Make a copy of the original before editing" means, that Audacity actually copies the entire audio file that you imported in to its project. The advantage of choosing to make a copy of the original is that you avoid trouble, should anything in the original file change. The second way is to use the original imported audio. Audacity will now read the imported file once and simply create the graphics overviews. The original file is only used for playback. For example, should you accidentally delete the original file, you're lost.

THINGS TO REMEMBER

Organize your folders before beginning your work One clip per track A clip is simply a piece of audio material. Imported, recorded, one track can only carry one piece of audio at a time and will always be one continuous piece of audio. Edit/Duplicate will not create a new audio file Audacity always records to a new track The UNDO function. You can undo/redo stuff as many times as you like.

RECORDING WITH AUDACITY Make sure your playback and recording device are set. Click on the red Record button to begin recording Click on the blue Pause button to pause the recording. Press it again to continue Click on the yellow Stop button to cease recording. The cursor will return to its previous position, before the recording was started.

IMPORT AUDIO... CTRL +SHIFT + I

Launches a file selection window where you can choose to import one or more audio files into the current Audacity project. The file(s) will always be added as a new track to the project. This lets you mix two or more files together.

NEW CTRL + N

Creates a new and empty project window to start working on new or imported Tracks. This new work environment can then be saved as an Audacity Project File (.AUP) for easy and full retrieval of its contents via the Save Project or Save Project As... File menu functions.

OPEN... CTRL + O

Launches a file selection window where you can: - Open one or more Audacity Project files (.aup) or - Import one or more audio files - If an empty project window exists, that window will be

used for the first project or imported file. Any subsequent opened projects or audio files will create new project windows.

If instead you want to add the content of an audio file into a project that already contains one or more audio tracks (for example, to mix two audio files together), use File > Import > Audio... instead.

CLOSE CTRL + W

Closes the current project window, prompting you to save your work if you haven't saved.

SAVE PROJECT CTRL + S

Saves the current Audacity Project. An AUP project file is created, along with a folder with the same name as the AUP file that contains the project's audio data. For example, if you save a project as corpus.aup, a folder called corpus_data will contain the audio data. The audio is saved in the lossless AU format.

EXPORT... CTRL + SHIFT + E

Exports the current Audacity project as an audio file format that can be read by other programs. If there are multiple tracks in your project, they will be automatically mixed in the exported data. For more information about mixing, see Mix and Render on the Tracks Menu. You can choose the exported file format and settings from the File Export Dialog. ! For all types of audio export including Export Selection and Export Multiple, tracks that are muted are not exported !

EXPORT SELECTION…

This is the same as Export, but it only exports the part of the project that is selected. For example, to save a small piece of a long track as a separate file, just select that audio.

AUDACITY SHORTCUTS - FILE MENU

Action Shortcut Description

New CTRL + N

Creates a new and empty project window to start working on new or imported Tracks

Open CTRL + O

Presents you with a standard dialog box where you can select either audio files or an Audacity Project file to open

Save project CTRL + S

Saves the current Audacity project .AUP file

Close CTRL + W

Closes the current project window, prompting you to save your work if you haven't saved

Import > Audio CTRL + SHIFT + I

Similar to 'Open', except that the file is added as a new track to your existing project

Export CTRL + SHIFT + E

Exports to an audio file

Exit CTRL + Q Closes all project windows and exits Audacity. If there are any unsaved changes to your project, Audacity will ask if you want to save them

Menu items assignable in Keyboard Preferences

AUDACITY - TRANSPORT MENU


Play/Stop SPACE Toggles playback on and off. When stopped, it is the same as the Play command. When playing, it is the same as the Stop command

Record R Recording begins on a new track at either the current cursor location or at the beginning of the current selection

Short Seek Left during Playback

LEFT or , Skips the playback cursor back one second can be set in the Seek Time When Playing section of Playback Preferences

Short Seek Right during Playback

RIGHT or . Skips the playback cursor forward one second

DEVICE TOOLBAR

Action Description 1. Change audio host Displays the Select Audio Host dialog for choosing the

particular interface with which Audacity communicates with your chosen playback and recording devices

2. Change output device Displays the Select Output Device dialog for choosing the playback device, but only if the "Output Device" drop-down menu in Device Toolbar has entries for devices. Otherwise, an output error message will be displayed

3. Change input device Displays the Select Input Device dialog for choosing the recording device, but only if the "Input Device" drop-down menu in Device Toolbar has entries for devices. Otherwise, an input error message will be displayed

4. Change input channels Displays the Select Input Channels dialog for choosing the number of channels to be recorded by the chosen recording device

1 2 3

AUDACITY SHORTCUTS - EDIT MENU


Undo CTRL + Z Undoes the most recent editing action

Redo CTRL + Y Redoes the most recently undone editing action

Copy CTRL + C Copies the selected audio data to the clipboard without removing it from the project

Paste CTRL + V

Inserts whatever is on the clipboard at the position of the selection cursor in the project, replacing whatever audio data is currently selected, if any

Find Zero Crossings Z Moves the edges of a selection region (or the cursor position) slightly so they are at a rising zero crossing point

Move Cursor to Track Start

J Moves the cursor to the start of the selected track.

Move Cursor to Track End

K Moves the cursor to the end of the selected track

AUDACITY SHORTCUTS - VIEW MENU


Zoom in CTRL + 1 Zooms in on the horizontal axis of the audio, displaying more detail over a shorter length of time

Zoom Normal CTRL + 2 Zooms to the default view, which displays about one inch per second

Zoom Out CTRL + 3 Zooms out, displaying less detail over a greater length of time

Fit in Window CTRL + F Zooms out until the entire project just fits in the project window

AUDACITY SHORTCUTS - TOOLS TOOLBAR

Action Shortcut Description Selection Tool

F1 Chooses Selection tool

Envelope Tool F2 Chooses Envelope tool

Draw Tool

F3 Chooses Draw tool

Zoom Tool

F4 Chooses Zoom tool

Time Shift Tool F5 Chooses Time Shift tool

Multi-Tool F6 Chooses Multi-tool

Next Tool

D Cycles forwards through the tools, starting from the currently selected tool

Previous Tool A Cycles backwards through the tools, starting from the currently selected tool



Remove Audio or

Labels > Cut

CTRL + X Removes the selected audio data and/or labels and places these on the clipboard. By default, any audio or labels to right of the selection are shifted to the left

Remove Audio or Labels > Delete

CTRL + K Removes the selected audio data and/or labels without copying these to the clipboard. By default, any audio or labels to right of the selection are shifted to the left. Closes the current project window, prompting you to save your work if you haven't saved

Remove Audio or

Labels > Split Cut

CTRL + ALT+ X Same as Cut, but none of the audio data or labels to right of the selection are shifted.

AUDACITY TRACKS MENU


Add Label at Selection

CTRL + B

Creates a new, empty label at the cursor or at the selection region

Add Label at Playback Position

CTRL + M Creates a new, empty label at the current playback or recording position

Labels can be used for textual annotation



Remove Audio or

Labels > Split Delete

CTRL + ALT+ K

Same as Delete, but none of the audio data or labels to right of the selection are shifted

Remove Audio or

Labels > Cut

CTRL + X Removes the selected audio data and/or labels and places these on the clipboard. By default, any audio or labels to right of the selection are shifted to the left

Remove Audio or

Labels > Delete

CTRL + K Removes the selected audio data and/or labels without copying these to the clipboard. By default, any audio or labels to right of the selection are shifted to the left. Closes the current project window, prompting you to save your work if you haven't saved

AUDACITY SHORTCUTS VIEW MENU Show Clipping Click this menu item to turn display of clipped samples on and off - a check mark in the item indicates that display is turned on. A clipped sample is one that is outside the volume envelope defined by +1.0 to - 1.0 (this is the vertical scale to left of the waveform when default Waveform view is enabled, as shown below). Clipped samples are bad, as they cause distortion and lost audio information - avoid them where possible, for example by setting the recording level correctly.

-Create a folder for the work For example 201408CollectionName -Open Audacity check preferences

« Save project as » giving it the box’s number (for example TAPEBAND453)

BRIEF CHECK LIST

Clean the player, clean the media if necessary

Digitize

Save the project

Start monitoring the input

Select the waveform and “export as wave” in the right folder, still with the box’s number

- Inventory your digitization on the inventory notebook (unique accession number) - Name the wave with an unique inventory number according to the inventory notebook (ex TBC564) - Copy the wave file in the folder SoundDigitizedWaveTBC and in the different hard disks. Never use for work the file inSoundDigitizedTBC.

-Export the wave file as MP3 to the folderSoundsDisseminationTBC

The room

The computer

=

The case

Computers internal hard drive D:

=

One card

One folder

=

CHECK LIST (SUMMARY)

-  Photocopy the different elements about the archive (cards, sheets in the file… everything which contextualize the document) and organize the inventory;

-  On the digital station, create a folder for the work (For example: 201408CollectionName); -  Open Audacity and "Save the project as … " (For example: TAPEBAND45); -  Switch the player off, clean it and clean the media if necessary; -  Open the tape box and check the accession number is the same on the box and on the tape; -  Ensure that the accession is on top when placing the reel on the player; -  Start monitoring the input and digitize in order to check the levels; -  Save the file; -  Inventory your digitization on the inventory notebook and give a unique serial number to your

file; -  Select the waveform and “export as wave” in the right folder, still with the same name; -  Rename the wave with the unique serial number according to the inventory notebook (ex

TBC564); -  Copy the wave file in the folder SoundDigitizedTBC and on the different hard disks. Never

work directly with the file in SoundDigitizedTBC; -  Export the wave file as MP3 using Cdex or Audacity (192Kbps if exporting with Audacity) to

the folder SoundDisseminationTBC; -  Catalogue your file in Excel; -  Write the digitization date on all corresponding sheets and cards.

Any questions ? Any questions ?

Crédits

Slide 1 and 8: Sound archives Marceau Gast, MMSH, photogr. Laure Principaud, janvier 2010. Slide 1: Digitizing sound, MMSH, photogr. Serge Mercier, 2012. Slides 1 and more : July 12, 1967 in northwest Burundi, Sekere, Emile Mworoha and Jean-Pierre Chrétien Slide 11: Introduction to Archival Terminology by Maygene F. Daniels (1984) http://www.archives.gov/research/alic/reference/archives-resources/terminology.html Slides about Audacity: wikipedia Slide 18: Sharat Ganapati, Hands Clapping at the Game, 2006 (CC BY 2.0), https://www.flickr.com/photos/frozenchipmunk/186912531 Slide 49 Nationaal Archief Follow Opname van een hoorspel / Recording a radio play, 1949, No known copyright restrictions

Archiving and disseminating sound archives – 3. Analysis and treatment of sound data

Education

Transcript of Archiving and disseminating sound archives – 3. Analysis and treatment of sound data