With budget limitations, it appears that oral histories are being recorded with little thought to their long-term preservation. While this appears to have been the case in the past as well, with purchasing agents buying the cheapest white-box tape that they could find, continuing this into the digital age needs to be reconsidered.
The cost savings in using bargain-basement digital speech recorders are offset by the labour required to reformat these files upon their receipt by an archive and also the fidelity of the recording suffers, and with fidelity, intelligibility also suffers.
DSS was an industry standard agreed upon by Olympus, Grundig, and Philips in 1994. Like the compact cassette and the micro cassette, this format was developed for dictation. My perspective is that oral history is more than dictation, although the DSS format claims no loss when compressing a WAV file. It includes nuances and other cues that are often lost in the dictation-only environment. My personal preference is for stereo recording of oral histories, but that is certainly not universally accepted.
While Olympus claims that DSS offers future-proof access to recordings, it becomes another standard that must be maintained and should converters become unavailable for new platforms, then the entire archive will need to be reformatted at that point.
The oral history resources I have listed here are generally looking at archiving WAV files. CD-quality WAV files should be more than adequate for oral histories, but that is a far different quality level from DSS.
There are a variety of software packages that will convert from DSS to WAV files. Switch (a great free utility) will convert SP mode, unprotected DSS files and a host of other file formats to WAV (and MP3). Olympus has a player as well.
The decision to archive in WAV or MP3 is tough. While concatenation of compression (stringing two compression cycles together, end-to-end) is considered poor form, if the MP3 file is not too heavily compressed, then there is saving vs. archiving in WAV if the program material originated in another compressed format. Tests with program material and critical listening should be undertaken prior to accepting this. If the decoded DSS or other compressed file is stored as a WAV file, then nothing is lost. The amount lost with (for example) a well-compressed 96 kb/s mono MP3 file at 44.1 ks/s will be low.
Archiving WMA files is another question. As long as Windows Media Player is available, these files should be playable. I’m not sure about the cross-platform compatibility of these files, so that is a negative at the start.
The best suggestion is to archive as few different formats going forward as possible, and then make sure at each major upgrade of software that these files will all be accessible.
A comment:
Your readers may also be interested in the Acappella Conference Audio Recorder and Playback Assistant which improves the reliability of transcripts and the time taken to produce them by recording in CD quality audio and telling the typist during transcription the name of the person speaking.
http://audiorecorder.wordpress.com
[Blogowner\’s comments: This was posted by someone who appears affiliated with the company. We have no experience positive or negative with this system. It appears capable and I released the comment to the blog but am adding this note that this is not an endorsement, but you should review it as a possibility and see if it meets your needs.]