Hello, Steph.
I recently recorded over 40 hours of field interviews using the Olympus
DM-1 digital recorder, with an external Sony ECM-F01 microphone. I found
this system works exceptionally well, even with significant background
noise. With a 128M flash memory card, I can record up to 22 hours of
audio without uploading to the computer. The whole kit (recorder and
microphone) fits easily into a pouch smaller than a Walkman. After each
interview, I uploaded via USB, which takes just a few seconds per hour
of interview.
I like the DM-1 because it s exceptionally portable (about the size
of a cell phone), very frugal with batteries (I got at least 10 hours
per set of new batteries), and also supports MP3 files (for when I m
not strictly working). The proprietary file format (.dss) is so small
that I ve been able to regularly archive my entire interview files
onto a single CD-ROM, with room to spare (this represents about 40
regular audio CDs ). An extra bonus is that the smaller .dss format
allowed me to email short (1-3 minute) audio files back home to my
girlfriend s computer, where I also installed the playback software.
The ECM-F01 is about the size of a credit card and runs around 120
hours between battery changes. It has a feature that boosts mid-range
sound and filters out background noise. I ve gotten very good quality
recordings with it, even with in less-than-ideal conditions, like an
outdoor cafe or a park bench.
You can find the DM-1 for under US$200 these days; 128M memory cards
are around US$40-50. The ECM-F01 is available for around US$65
I have tried using the Olympus interface to create transcriptions
myself, but have found the Transana software (available free from the
University of Wisconsin) is much easier to use. The program is capable
of processing both .wav and .mpg files (it was originally developed to
aid video transcription) and includes an internal file management
system. To shift the DM-1 files into Transana, I use the Olympus
interface to convert the .dss files into .wav format.
Transana is the easiest program I've found to make transcriptions, as
it is especially easy to jump forward and back within the audio file,
without taking your hands from the keyboard. I am transcribing English
language interviews in a little more than double the duration of the
interview. In other words, it takes me about 2-3 hours to transcribe 60
minutes of audio.
Note that you still must type out the transcriptions yourself (speech
recognition of random sources is still several years away).
Hope this helps.
Ken Cousins
Harrison Program on the Future Global Agenda
Department of Government and Politics
University of Maryland, College Park
T: (301) 405-4133
C: (301) 758-4490
F: (301) 314-7619
[log in to unmask]
"The important thing is not to stop questioning.
Curiosity has its own reason for existing."
Albert Einstein
