Saving Audio File

Is there anyway to record and save audio files in a SuperLab experiment? I want to run experiments in which the subjects’ response is spoken.

Unfortunately not.

This seems like a major limitation to me and something that would be very useful to consumers.
Does anyone know a way to dump a document that uses dication software (e.g., dragonspeak) into Superlab so real time audio output could be displayed?

An old trick

Something we used to do in the old days was to use a multi-track tape recorder to record the audio or other waveforms on one or more tracks, and then use an fsk or other encoder–basically, an old (e.g., 300 baud, 1200 baud–modem) to record the events on another track. A digital recorder can of course do this much more neatly than that, but the general idea would remain the same. Whether this could be run on the same computer that Superlab is running on is of course another question. On Macs, the ever-popular Audio Hijack application does come to mind as a possibility.

I can think of two general ways this could be done with Superlab. If you don’t need audio output during the experiment, a low tech method would be to have Superlab produce very short audio-frequency beeps at certain frequenies when certain events occur on one input of a stereo recorder (or even another computer); the other input would be fed from the subject’s mic. The resulting waveform could then be analyzed with any stereo waveform editor.

The higher tech way would involve a multitrack digital recorder and Superlab’s digital output capabilities; this now becomes somewhat similar to how evoked potentials or something along those lines might be recorded, but with an audio waveform. Much higher-level analysis software exists that could probably be used for this option.

An intermediate method, if you do need to use audio output in the experiment, would to use Superlab’s digital output to produce the beeps with some kind of external circuit (there are many ways to do it, including classic FSK).

Greg Shenaut

thanks Greg

…If I understand your ideas correctly, you are talking about producing audio feedback…I am wondering about combining the output of one program that essentially takes dictation.

Can a program like this be linked with or embedded into Superlab?
I am on day one of trying Superlab so forgive the naivete.


No, not audio feedback

I’m talking about using audio frequencies to encode events and storing them along with the subject verabl responses on a stereo tape. The subject wouldn’t hear them, they’d be fed directly into one of the audio inputs. The purpose is so that you can get timing and response code information from an audio tape.

Here’s a way to think about it: have you ever seen one of those spy thrillers where someone uses a taperecorder to record touchtones, and then later play back the taperecorder to dial the same number? Imagine that you’ve recorded very short samples of the 10 standard touchtones, and programmed Superlab to produce certain ones at key moments of the experiment through the computer’s audio output, which is connected to one input of a multitrack recorder. The subject makes responses into a mic which is used to trigger a voice key, and also fed into another input on the multitrack recorder. In the multitrack recording, you will have arbitrary timing and event codes (up to ten different codes, which could be combined to produce complex sequences) and also everything the subject said. This can all be analyzed later with a waveform editor or (obviously) more specialized software.

That’s the basic idea, I guess it all depends on what you need to do with the recordings.