Stand-Alone Recorder

This will get more exotic as we go, but this is the first sound test of an Olympus WS-823 stand-alone recorder. It’s the cheapest one that can be set for WAV output. $120 USD

It did need some technical help, but surprisingly little. No noise reduction.

Recorded in my quiet third bedroom.

Screen Shot 2015-08-24 at 22.18.36.png

Hi Koz

I like Olympus because they make accessible voice recorders (with the so called “Voice Guide”)
However, the pro recorders that offer high quality wave recordings are not accessible.
It really annoys me that these manufacturers are of the opinion that visual impaired people don’t need high quality audio recordings.

Your sample sounds great. Have fun with your device.

Best regards

That’s a very clean recording . Is that using the internal microphone[s] , or an external one ?.

There are some external mics available specifically for that device which have [non-digital] noise-cancelling for use in noisy environments [ only $15 ], shown here much bigger than life-size …

Stand-alone everything. One candy bar jammed into a mic stand. Pictures to follow. I have a clothespin/expanding microphone clip on my mic stand and I clipped the recorder into that—along the bottom so it didn’t interfere with the acoustic management at the top.

I could not be happier that it doesn’t need noise reduction. Somebody was paying attention. I’m getting used to how the tools work. I recorded a meeting with the most sensitive “conference” settings and after carefully listening to the partially overloaded sound file, I discovered it was cheerfully recording the idle conversations in a room three offices down the hall.

This is me learning how to use the settings.

It also has a distinct character or quality change between end-on and facing the readout. I picked facing the readout because that setting sounds most like me. I suspect End-On is intended because that orientation is louder.

I’m deliriously happy the sound doesn’t need de-essing (harumph, caff, caff).

And yes, vision-free users need not apply. I have trouble following the menus and my natural clear vision is six inches. I think it’s a setting, but it decides to shut down the backlight at the exact instant I make the final setting decision. Son of a !@#$%.

Again, I believe this is the lowest-end Personal Recorder that will record and export WAV 44100, 16-bit, Stereo. It will also produce a 256 quality MP3, itself not too shabby. And to be crystal clear, I recorded it between cars, at night, in my soundproof bedroom. I really need to figure out a way to “turn the window off” when I do this.


Another note. I got a boost to do this from someone who started recording an audiobook (one of many) by squatting in a hotel room closet full of blankets and presenting into a small recorder. Seemingly well, too.

I don’t remember the conditions or who. Behold, the right-brain engineer.


Does it have a “real” input level control? (can you record loud things without clipping?) It would be interesting to know what sort of job it can make of recording music (recording a live band). If it doesn’t clip it could be a good option for recording rehearsals.

The “manual” recording setting is affected by the generic +/- control on the front. As is common with these things, it never goes up high enough, so I have the setting jammed all the way up and I can’t hit -6dB meter with my voice. But I can hit -12, so that will have to do. I have no idea where in the chain volume management happens, but I do know that there is direction and pattern management and they warn you that battery life is affected if you elect to use those tools.

The down side of well-behaved MicPre is usually the inability to control it.

I have no idea how it would respond to energetic Fender Bass exposure. Worse, LAX closed “my” runway, so I can’t use that for sound testing until they finish repairs and upgrades.

Say goodbye until sometime in October. Two-Four-Left is still active, but it doesn’t give the sensation of Air France landing over your Double Double with Animal Fries and ice tea.


That’s another decision. I created that clip by Stereo > Mono mixing. I could also have used either of the two directions. I think the noise gets better when I do it this way. By what? Square root of something something?

I did some of the experiments in stereo conference recording at work. Very Highly Recommended. I demonstrated I could hear and understand three different conversations going on at the same time by presenting a very high quality stereo recording of the meeting. Engineering thought I was nuts, but the production people were delighted.


Square root of 1 divided by two = 6 dB at hard left or right.
Again, Center Isolation does a better job in this regard, minus 120 dB at least.
Pan and mix&render to have a certain band in the middle and apply then the center isolation (e.g. for the case below).


Center Isolation does a better job in this regard, minus 120 dB at least.

May not be an unqualified joy. That depends on me being in the exact center of the microphone during the presentation. If I’m not, that portion of the voice direction error will drop out of center isolation giving me a swimmy effect. As it is, Stereo to Mono doesn’t sound exactly like the full stereo version, but it’s so close I didn’t go back and inspect it.

Given a bought and paid for song that was originally shot with the lead singer on one single microphone, then yes, certainly. The lead singer was never in stereo.

Maybe a neck brace…


OK, here’s a poor example.
I’ve used a Rode NT 4, directly plugged into line-in - I’ll get my new audio interface this week, which will result in better quality.
I’ve applied a lot of noise reduction, thus the artefacts you hear.
Anyways, as you can hear, the TV plays on the left side.

I then appended the isolated version - with “Strength” at minimum, i.e. 0.02.
This removes only the very extremes at hard left and right. In other words, you could move within a certain angle while speaking.

It’s generally like this:

  • Strength < 1.0 - the isolated center has the form of a ‘U’ upside down.
  • Strength = 1.0 - the isolated center has the form of a ‘V’ upside down.
  • Strength > 1.0 - the isolated center has the form of a ‘T’ upside down and left right movement during recording will be punished.

    Mixing down to mono reduces the TV sound by another ~ 12 dB.

Again, please don’t mind the quality.
Perhaps, you could provide a sample shot with the Olympus with some distracting noise on the sides.
Of course, if the noise is only on one side, it would be enough to split the track and just use the better channel.


split the track and just use the better channel.

Losing the goal and working too hard.

I have the original shoot here. I think.

Oh. Shucks. I can’t post to my web page for a while. They’re moving it. I need to make other arrangements.

Just listening to the test again there is a 2 to 3 dB shift between channel volumes and it doesn’t stay the same direction.

This recorder has a setting that causes the Cone Of Acceptance to expand or tighten. In fully tight, you can pick out a performer in a noisy room (it says here). I use the widest possible setting for conference recording. I think I used the default setting (middle) for the voice test.

As we go.