Home » Articles » Using a Digital Recorder for a Ghost Box Session

Using a Digital Recorder for a Ghost Box Session

Proper Use of your Digital Recorder for a Ghost Box Session

by Tim Woolworth


Are You Using a Digital Recorder for a Ghost Box Session Properly?

When we conduct ghost box sessions, typically the only evidence we have as a result is the audio we have recorded – and usually this audio is recorded using a standard digital voice recorder. On rare occasions, ghost box users may record their ghost box sessions on video as well, but this is far from the norm as of now. For those of us involved in Ghost Box sessions (or any other form of Instrumental Transcommunication for that matter), we have to realize that this paranormal field is still in its infancy. Because of this often overlooked fact, there is a duty that we must fulfill every time we conduct a ghost box session. We must do our best to not only record our ghost box sessions the best way we know how, but then to ultimately preserve our audio evidence for the future. ITC Voices will preserve the audio that we have recorded for future ghost box users to listen to, but recording audio with the most integrity is upon us now as ghost box users.  How to achieve optimal audio integrity is paramount, and to do this, it is necessary to understand how you are using a digital recorder for a ghost box session makes all the difference in recorded ghost box audio.

I have spent a good portion of my adult life working professionally with sound. I am drawing off of my experience and knowledge as both a studio and live sound engineer to offer you a basic understanding of how your digital recorder actually achieves its most basic function: recording! To get the most out of your recorder (and hence the best audio quality for your ghost box sessions) you should absolutely read this article. There are quite a few technical concepts in here, but I will make them as simple as I possibly can for you to understand them. I hope the information contained herein will allow you to use you digital recorder more effectively so ultimate achieiving and maintaining audio integrity can occur while recording your ghost box sessions.

Recording in the field or at home is easy when you are doing an EVP session. Typically you find a place to sit still and either hold the recorder in your hand or set it on the floor and conduct your session. If you examined the way you conduct a ghost box session, I am willing to bet that you have found yourself doing the same thing with your digital recorder more than once. Unfortunately, utilizing a digital recorder this way during an ITC session will not give you good, clear quality audio.

Before I get in to ghost box recording techniques, I need to explain to you the fundamentals of digital recording first.

In the new age of digital recording, the audio sampling rate is where everything starts. The higher quality the recorder, allows you to use a better the sampling rate. A cheap digital recorder may have a sampling rate of only 8 kHz or 16 kHz, whereas a good recorder may have a sampling rate of up to 44.1 kHz or higher. The only way to find out the sampling rate of your recorder is to look at that wonderful technical specifications page from its manual or website. You may be surprised at what you see! That Olympus digital recorder you picked up for forty dollars will not even come close to giving you the recording quality for ghost box sessions or EVP that is necessary; their are many reasons why this is, but the most important is the sampling rates that are available to record with.

But what does sampling rate mean?

It means that for every second of audio that you record into your digital recorder, a pre-determined number of samples are taken of that second long piece of audio. This is the sampling rate. The cheap recorder, or even a great recorder with a wrong setting, may only have a sampling rate of 8 kHz. This means that for every second you record of audio during a ghost box session, your recorder slices that second into eight thousand little bits of digital information called samples. This may sound like quite a lot of work for a second and that 1/8000th of a second should be all you need to record great quality audio. Take this into consideration though: the higher end recorders will slice that same one second of audio into forty-four thousand one hundred samples per second. For a higher sampling rate, you sacrifice available recording time.

You have probably noticed that your digital recorder has different settings to record with. It may be as simple as HQ for high quality or SQ for standard quality, the settings are different for each type of recorder. Because a higher sampling rate records so much more audio, it allows you less time to record on your device during an instrumental transcommunication session. So until now, you may have been using a standard quality setting in order to get the most available time to record with during a session. This is a big NO NO! Always use the highest quality setting in order to get the highest sampling rate available to record your ghost box session. You may have less time available to record with, but the quality will be so much better that it will be worth it. For an example, the sampling rate is the equivalent to recording on a VCR tape. If you ever had a VCR, you will remember that it had three setting to record on: SP (standard play), LP (long play), or EP (extended play). The SP would always record with the highest quality, but a tape was limited to recording only two hours. A VCR tape set to EP would give you upwards of six hours of recording time, but the picture would be much grainier and may have some static lines going through it at times. The same is true of sampling rate, but instead of magnetically charging particles on analog tape, you are slicing the audio into thousands of little samples.

You may be saying right now that even with a smaller sampling rate it still sounds good and I can record a lot more audio too! After all, splitting a second into eight thousand pieces sounds pretty darn impressive. This is true, it is. But, as with everything, the devil is in the details. Now let’s say you have the best ghost box ITC session of your life and you heard everything in real time and received real time responses. You immediately download it to your computer and then realize that upon playback, that there is even more that you missed because some responses are uttered really rapid and you need to slow it down in order to make sense of it (this very common in ghost box communcation). So you load it into your favorite audio editor and try to stretch out a section of rapid syllables that you think may have a valid answer to one of the questions you posed to yours friends from the Other Side.

If you have ever done this before, you know what the result will be. Noise. If you have only 8 thousand bits per second to work with, when you stretch them out, the gaps between these 8 thousand bits become much more apparent and all of a sudden your recording has a robotic sound as your audio editor tries to resample the audio and stretch it out to fill in these ever wideing gaps. You would be surprised, but I have listened to many clips online that have had this effect done to it and I feel that they are wholly unusable because the resultant audio from this processing is simply unstable from being laced with digital artifacts caused by both your audio recorder and audio editing software. If you had forty-four thousand odd bits to work with, there would be fewer gaps in the audio and therefore less digital artifacts after effects processing.

It should be noted than ANY time you use processing like stretching, speeding up, noise reducing, or reversing audio in an audio editor, the audio you have will be laced with digital artifacts and is therefore unreliable. I am of the opinion that if you have to use any of these effects to make sense of a piece of audio, the audio may provide you some sense of edification and you may understand it the answer as it becomes clearer than it was before, but as a piece of evidence, it is circumspect at best and you should not present it as evidence. In a digital world, you cannot ever record everything that the ear hears in one second of audio. As technology progresses, that one second wil be represented by more and more samples, but capturing the whole essence of the analog sound is a mathematical impossibility in any digital recording whether it be in audio, video, or the photographic medium.

So now that you have a firm grasp of what the sampling rate is on your digital recorder and how you will use it for your next ghost box session, it’s time to consider what frequencies are you actually recording.

You may be saying “huh?” after that one. Let me explain.

All humans can hear audio within the range of 20 Hz to 20 kHz. This means that if your hearing is not damaged at all, you can hear the bassiest sounds that consist of the lowest, and therefore the largest, sound waves. These waves are so big that only twenty of them happen per second. The unit Hertz measures the amount of waves in one second, so twenty Hertz, or 20 Hz, means twenty waves per second. Each Hertz measurement is a separate frequency. The highest end of sound that we can hear has over twenty thousand of these waves per second so they are really tiny in comparison to the bass waves. The average middle-aged adult has a hearing range of about 28 Hz – 15 kHz.

So every digital recorder should record you ghost box session with the full range of 20 Hz – 20,000 kHz, right? If you choose a low sampling rate like 8 kHz, the frequencies it can record is also curtailed significantly. Surprise. I went to the Sony site and looked up the sampling rate of some of their recorders. A typical recorder that utilizes a 44.1 kHz sampling rate records audio frequencies up to 20 kHz, or the full range of human hearing. But, this same recorder (which is a very high end one I might add) with only 8 kHz selected, only has the ability to record audio frequencies up to 3.4 kHz; this means that if you have the lowest sampling rate selected, not only will you record less audio per second, but you will only records 1/6 of the available frequencies that are being spit out by your ghost box. Imagine that you only heard one out of every six words in a conversation; you would feel that you are missing something very important. If you are missing almost seventeen thousand wavelengths, there is a lot of audio that is not present. This missing audio can mean the difference between hearing one word or a different word entirely.

So what does all this sampling stuff boil down to? More is better! If you have a cheap recorder that can only record at 8 kHz, then it is certainly time to upgrade if using a digital recorder is your medium of choice. Anytime you record at a low sampling rate your resulting audio will not have the integrity that it needs to have in order to be examined without bias. If you use a low sampling rate and then present the audio you have as evidence, I would liken it to going to the movie theater and wearing dark sunglasses during the film: it is the same film, but the quality of the experience is less than desirable because you would miss out on all the carefully chosen colors and lighting schemes.

One last thing you need to know about digital recorders is something called bit depth. If you look at a commercial compact disc, you will see that it will say something like 44.1 kHz/ 16 bit which is the commercial audio standard, also known as red book compliant audio. You already know what the sampling rate is, so what is this 16 bit all about? A bit described at its most basic is the level of accuracy at which the sampling rate occurs. Obviously, recording those thousand os slices of audio during a ghost box session with more accuracy will give you far superior audio than if you recorded to same piece of evidence with less available bits and sampling rates.

So how does a bit work? When audio comes in through a microphone or a cable, it is considered an analog signal. In order for this analog signal to work in a digital device, it has to go through an analog to digital (or AD) converter. This converter does exactly that, it takes the analog audio coming in through the mic or line in and converts it into a digital language l that can be understood by your recorder. Luckily most digital recorders these days only go through one stage of digital conversion because the digital bits inside the recordeer be transferred to a computer via a USB cable; this is digital to digital transferrance because the cmputer uses the same digital language as your computer.

Most recorders have a bit depth rate of 16 bits, but there are some (wholly dependent upon quality) that can supply only 8 bits or up to 24 bits of recording quality. These bits are a measurement of how many binary numbers are in a sample. Binary code is the language of digital devices. Everything that is digital has a binary number which is constructed by stringing together lines of 0’s and 1’s.

I know this is getting confusing, but bear with me a little longer and I will clear this up.

So we have this one slice of a second of audio from a ghost box session. The question then becomes how does this one slice get converted in relation to the thousands of other slices taken every second. It becomes a question of accuracy.

In an 8 bit system, each sound slice and the wavelength (Hz) it contains is placed in relation to 1 of 256 binary locations. If the size of the wavelength does not fit exactly onto one of these pre-determined binary locations, it is moved to the closet binary number that represents the wave’s amplitude. This again falsifies a bit (pun intended) of your audio because the digital representation of that little slice of audio doesn’t have a snowball’s chance in hell of being accurately converted because of the exceedingly limited number of binary characters it is converted with. But, if you have a 16 bit recorder, that one slice of audio being accurately represented with a binary string become exponentially greater. In a 16 bit system there are an astounding 65,536 binary numbers to represent the amplitude of each sound slice. Chances are the bulk of the audio will be converted more accurately. The highest end digital recorders give you 24 bit recording with 48.1 kHz sampling. In a 24 bit analog to digital converter, there are over 16 million binary numbers to pair your audio slice with which lends itself to astounding accuracy. I would not be surprised that with a high sampling rate and 24 bits recording, you may be able to record a ghost fart and have it sampled accurately. Consider this: we all have heard that Hi-Defintion television is far superior and we can visually see that it is much better that standard digital signals on your television. Why? Digital televsion uses only 16 bits (remember that is equal to 65,536 binary numbers). Hi-Definition televsion can utlize up to 48 bits of depth which provides over one trillion binary values to represent an analog sample.

Believe it or not, I once met a guy who wanted to ghost hunt with me. I asked him what gear he brought with him and what astounded me most was his audio recorder: it was 4 bit and 8 kHz! He told me that wherever he used this recorder he recorded EVP’s. He thought he was a ghost magnet with a special connection to the other side. I had to chuckle inside because the bit depth and sampling rate were so low that nothing that recorder EVER recorded could be trusted because of its mechanical limitations. I did not even bother trying to explain it to him because he thought he was the cat’s pajamas. He later sent me a clip of the “voices” he recorded and it was pure artifact noise with static!

There is one last thing I will go over about recording quality audio in an ITC ghost box session: volume. Simply put, louder (but not too loud) is better.

When you use your recorder, chances are you use it like you would in an EVP session. You probably hold it in your hand or set it down somewhere close to the speaker. If you are doing this on a ghost hunt, there is a very good chance that you are doing this is the dark. This can be VERY problematic because it undermines the quality of bit depth.

On all but the cheapest of digital recorders out there, the recorder will reproduce a digital representation of the volume you are recording at. When using digital, the very top of this volume meter is labeled “0” and everything underneath it is a negative number. Don’t worry; I am not going to talk about negative numbers in math here! But I will dredge up a concept I briefly discussed earlier, the amplitude of the wave. As mentioned, the amplitude of the wave is what is AD converters attempt to reproduce by using binary numbers; the amount of binary numbers available to represent the wavelength is reliant upon the bit depth, and to complete the chain, the bit depth is relaint upon the volume you record at.

The goal in digital recording is to get as close to the number “0” on the volume meter without going over it. At zero, the sound slice has the maximum amount of binary numbers available to try to represent it. I won’t drag out more numbers to confuse you even more, but basically for every 6 decibels you are down from 0, you lose 1 bit of audio quality. A standard 44.1 / 16 audio recorder has a range of 96 decibels (dB – a measure of loudness). So, when your digital recorders volume meter shows that you are only recording at half volume, take a wild guess what that does to your recording quality as you reduce that sixteen bit recorder to an 8 bit one unintentionally by not having your recorder placed in a spot for maximum volume efficiency?

The goal is always to get as close to 0 as possible without going over. If you peak the volume and go over 0, something called clipping occurs. A digital clip is a very unpleasant thing. When clipping occurs, the converter takes the sound wave and slices off every bit of it that goes beyond 0. If you have a hot sound source that spikes for two or three seconds, all of those waves will be cut off and assigned the highest binary value available. These waves then result in a glitch sound because you have 44.1 thousand slices all being assigned the same binary number for an extended period of time. The resulting sound is horrible.

Unfortunately, no ghost boxes on the market have limiters or compressors built in to control the volume of the lows and the high volume spikes. Your digital recorder cannot control these either. Ultimately, to get the best sampling rate and volume, you have to record into a computer with audio software that has pre-fader compressor/limiter software, or, use a digital recorder with an outside mixer and compressor/limiter. Obviously, in the field, neither of these set-ups will work so the volume you record at has to be monitored if you want a quality recording. This should make you think twice about recording in the dark!

I know some people who think that if they amplify the recorded low parts later in software they can make their audio sound better because it will be louder. Nothing could be further from the truth. Unfortunately, you only get one crack at recording digital audio and there is nothing you can do in audio software to make up for lost bit depth due to volume issues. You have probably experienced this yourself, if you have something recorded at half volume and you amplify it in an audio editor, what do you get? Double the noise and audio that degenerates.

I started this article by saying that we need to actively attempt to achieve the highest quality audio recordings in order to present our ITC ghost box sessions as credible evidence. Hopefully, if you made it this far, you will have learned that your digital recorder, and how you use it, means everything to the quality of audio you record.

As always, if you have any questions, feel free to email me: tim@itcvoices.org


Have Your Voice Heard! Leave a Reply.

This site uses Akismet to reduce spam. Learn how your comment data is processed.