- Audio Capturing Process (Encoding) - Also called
digitization, this is the process whereby an analog source consiting
of waveform with a given frequency, amplitude, and timber is converted
into a digital electronic waveform or file. The file is expressed by
two sets of binary values: sampling rate(see below)
and bit-depth(see below).
- Bit-depth - When digitizing audio the "bit-depth
refers to the size of the binary numbers assigned to describe the dynamic
value of each sample.(Beggs
& Thede, 43)" With higher bit-depth more subtle
nuances especially for softer dynamics and long decays are enhanced.
16-bit is standard, but some audio professionals are pushing for 20
or 24-bit encording.
- Compression - A process in mastering a sound file
that reduces spikes in sound that are much louder than adjacent sounds.
If these peaks aren't reduced, you can't effectively normalize a file
to get all the sound waves at optimal sound intensity. The peak limiter
process or the amplify/gain process in SoundEdit 16 is used to compress
sound files. 3dB to 9dB of compression is recommended in order to add
clarity and brilliance to the sound (Beggs
& Thede, 84).
- Compression Formats - Several such formats exist
that reduce audio files in order to make online file sharing, streaming,
and downloading more feasible. In the encoding process frequencies outside
the range of human hearing and other "non-essential" parts
of an audio file are eliminated. Also a file is optimized with codecs
(compression-decompression encoding algorithms) and once recieved by
the client reconstructed. Due to the Napster lawsuit, MP3 is perhaps
the most well known compression format. Other popular formats include:
Real Audio, Quick Time, Shockwave, LiquidAudio, Advanced Audio Coding,
and Windows Media Audio. Standard uncompressed sound files are either
AIFF or WAV files.
- Digital Audio - "The term applied to sound that
has been digitized into binary code. Sound waves are converted to digital
code by means of a circuit that assigns a value or binary set of numbers
that correspond to the shape of the incoming electrical wave or current"
(Beggs & Thede, 356).
- Downloading Audio - As opposed to streaming audio,
downloading audio means that a user can have a permanent copy of a file
stored on their computer's hardrive. Allowing files to be freely downloaded
from peer to peer networks led to the highly publicized Napster Recording
Industry Association of America (RIAA) lawsuit. Profits for the recording
industy have decreased since the introduction of MP3 technology, so
the RIAA will likely not overlook library digitization projects that
are infringing upon their copyrights.See copyright
section.
- Equalization - A process used to increase or decrease
the amplification (loudness) of a given range of frequencies. The result
is an adjusted and better overall sound - a sound that accoustically
matches other sound files and/or takes out frequencies that will be
distorted or lost when a file is encoded (Beggs
& Thede, 85).
- MP3 - Officially MPEG-1, Audio-Layer III, is a popularly
used compressed digital file developed by the Motion Picture Experts
Group. The file size of the original file is reduced by one-tenth. A
newer format, MP3PRO which reduces size even further without losing
quality has recently been released. This is a non-proprietary format
with fewer obvious "commercial" connections.
- Mastering - "Mastering refers to the process
of preparing a group of sound files (i.e., music tracks of a CD) for
final delivery in one uniform manner, medium, and level of quality.
This means that each sound file has little unwanted noise and shares
the same relative decibel (volume) levels and equalization characteristics
with the rest of the audio files"(Beggs
& Thede, 79). In a digital library project this is more
important when digitizing analog sources than in pre-mastered CDs, although
other digital content - CDs of student recitals - may need some mastering
before putting out on a website.
- Normalization - An audio mastering technique that
raises the overall amplitude (loudness) level of a sound file so that
it: 1) matches other sound files in a grouping, 2) "maximizes the
bit-depth information available by giving a sound a louder overall presence,"
and 3) helps prevent distortion when going from a stereo file into a
low-bandwidth mono-file (Beggs
& Thede, 81). Normalizing to 95%
or -3 below 0dB is recommended unless #3 is the case, in which you want
to normalize to approximately 70% of the input signal level.
- Quality of Audio - The higher the sampling rate and
bit-depth of the file, the better the quality though the larger the
file size. File size and quality must be balanced, while also taking
into consideration the capacity of both local and remote users, i.e.
their bandwidth or connection speed. The less compressed the file, the
higher the kbps, and the better the quality. Though each compression
format varies, for MP3 the following compression rates apply - 128 kbps
= CD quality, 96 kbps = near CD quality, and 64 kbps = FM radio quality.
- Sampling Rate - During the
encoding process the sampling rate (measured in Herz (Hz) or cycles
per second) determines how much of the sample is replicated in a given
frame of time. The higher the sampling rate and bit-depth the larger
and more "complex" the sound file. The sampling rate determines
the rate in kbps (kilobites per second) in which sound bits are delivered
from server to client. Engineers and conservators argue foor 192 MHz
and 24bit sampling (Brylawski,
63); however, sampling at 44 MHz, 16bit (standard CD) is
perhaps still an acceptable standard audio sampling rate, because it
is the lowest rate that accurately reproduces the highest audible frequencies.
Libraries sometimes over two sampling rates to accomodate users with
different Internet bandwidth connections.
- Streaming Audio - The ability to deliver sound in
digital bits or packets over a connection to a user, with additional
bits remaining in a buffer until needed and being discarded after played.
A server needs to be set to stream audio, yet this is a wonderful means
of delivery, because an audio file can be delivered this way with little
delay while making the illegal copying of files more difficult. The
illusion of seamless play relates to the buffering of packets before
playback and the simultaneous downloading of more as play continues.
Though one can have pseudo-streaming with a regular HTTP setup, new
Internet protocols like UDP & RTSP transmission are more reliable
and can deliver more streams.
|
