Recording samples in the digital age

by Eric Armstrong

For the past twenty-five years or so, the main technology for recording dialect samples (and various other snippets of voice) used by voice and speech professionals has been the audio cassette. In recent years, two recording technologies have appeared on the market which offer many features that go far beyond what a typical audio cassette recorder can do. MiniDisc (MD for short) and DAT (an acronym for Digital Audio Tape) are both well established standards in the communications and theatre sound worlds. If you're in the market for an upgrade, you work in the dialect coaching field or you plan to create your own samples for classes or for sale, some of these features may induce you to lay out the cash for a portable recorder.

In preparing this article, I spoke with Michael Barnes, who teaches voice and speech at Temple University, and freelancer & dialect author Gillian Lane-Plescia. Michael inherited a SONY DATman recorder through his department, and Gillian has been using her MD in combination with her old portable audio cassette recorder. We discussed the pro's and con's of both technologies in the context they know best, which is sampling dialects in the field for use in the professional theatre and in the classroom.

The greatest advantage of the two technologies is that the sound is recorded digitally, as opposed to the analog format used with an audio cassette recorder. They are formats that most of us are familiar with: Compact Discs (CDs) are digital, LPs are not. Wanting a simple explanation, I turned to the internet for help. The "DAT-heads FAQ" (to translate: Frequently Asked Questions, answered by lovers of DAT) provides us with some clarification:

"Sound consists of rapid pressure variations, called "waves", in a medium such as air. Sounds have traditionally been recorded, processed, and transmitted as electrical signals which have waveforms that are analogous to the waves in the air. This is where the term "analog" comes from.

Digital waveforms don't look anything like sound. They can only be used to represent the binary values 1 and 0. But a whole bunch of 1's and 0's can be strung together to represent any number. And a whole bunch of numbers can be strung together to represent the "image" of almost anything. Digital audio uses numbers to represent the image of an audio waveform. With enough binary digits (bits), the image can have such great resolution that it would be impossible to tell it apart from the original analog signal.

A digital audio system typically involves analog to digital conversion, storage of digital data, digital signal processing, and digital to analog conversion. There are many advantages and disadvantages to digital audio and they are often the subject of intense debate. In general, for a given level of fidelity, it is more economical to use digital than analog."

Digital has the advantage of not degrading in the editing process (i.e. the 0's and 1's stay that way). Digital also has better signal to noise ratio, so that there is less hiss on your recording, and it has a greater dynamic range - very important to the recording of music, though not so important to the recording of speech. Digital formats also tend to be more easily accessed much like skipping around on a CD player. Both have the disadvantage that you must copy your material to another format to share with those who need it, as so few people have MD or DAT. But if you do a lot of sampling, a digital recorder will make master recordings which will last longer and sound much better.