Once every 10-20 years, something revolutionary comes along. This is especially true in music. As of late, Artificial Intelligence (AI) is the new buzz-word. While often used as a mere marketing gimmick, AI is proving daily that it is the next game-changing technology. And now, Spectrelayer is here. This new service has nearly a decade of development & as much lead time. It’s time to let it out of the proverbial “lab”.
You know that over 70% of music sales are currently what is called “Older Genres”. Because those artists from the 70’s & 80’s are also the favorites of the post-“Boomer” generation, a unique sales window is open for about 10-20 more years. You see, the only thing better than selling a record once is to sell it twice. And the people who might buy it a 2nd time are those who purchased the original release. There is only (1) reason that somebody would do that: An audibly superior re-release of their favorite material becomes available. When companies began to sell digital re-releases of previously analog material – they realized the simple truth of what I’m talking about: Compact Discs replaced tape & vinyl with a roar at the cash-register! But what has the recording industry offered since the digital revolution became par? Remasters. A “Remaster” is where musical classics (art) are reissued after having been damaged with “loudness-war” mentality & relabeled advertising it. That’s not true every time – but it happens enough that it is now a stereotype. And such “remasters” along with “Hi-Res Digital”, MQx, “snake-oil 2.0”, etc. have damaged the reputation of the music industry. Spectrelayer is different – clearly, audibly superior: POSTmastering.
As the process developer, a polymath & autodidact; -I wish that Spectrelayer was a simple black box that I could pour a recording in on one side & have it emerge processed on the other with everything perfectly ready for return-submission. But in the world of quality art – that’s not how things work. Spectrelayer still requires a human with ears & expertise to review the content & make adjustments. And as anyone who works with AI will tell you: There are artifacts & that must occasionally be addressed. Just as sow’s ears cannot be easily made into proverbial silk-purses; -Neither can challenging audio be easily reconstructed without an engineer’s ears meticulously guiding the process. And because engineers who understand the workings of a DAW -AND- integrating such with proprietary technologies (such as AI) – are scarcer than honest politicians; – He who hesitates had better be prepared to wait in line once the proverbial trigger gets pulled. NOW is the day to begin your relationship with Spectrelayer as a client. Test drive us w. our free & discount offers!
Questions & Answers about the Service that is giving new life to recordings of great music!
Q: What Inspired the Spectrelayer Name?
A: The root term in the Spectrelayer name is: SPECTRE (spek-tər)- The badass European spelling for a term associated with a ghost or spirit (one usually deemed to be startling or attention grabbing). And this is the goal of the Spectrelayer process; -To seize & focus attention of the listener by remarkably improving the audio. But WHY? Because large sections of the population find extreme enjoyment in listening closely to music – not only compositional complexity but also the philosophical insight of the lyrics. To many people – music is a Spiritual experience of unparalleled beauty. Even 2000+ years ago, Rabbi Yeshua Ben Yoseph said that the rocks themselves could speak unspoken truths (Lk 19:40), & here we are – in a day when microprocessors (made of purified mineral crystals – rocks) are being used to render information & reveal hidden details! Spectrelayer is the proverbial “ghost in the machine” – being given a voice. And when you consider what systems like AI can bring to bear – wow! The original pronunciation of the term was intended to be “SPECTER+LAYER”. However, when concatenating the term “layer” after the “e” in “Spectre”, – reading convention tended to change the accent’s location & hence, the pronunciation. I eventually gave up trying to correct people & now accept either pronunciation when I hear it (although those who pronounce the term “SPECTER” followed by “LAYER” impress me insomuch that I know they’re paying attention to details).
Q: Why the use of colors in the word-logo; -& How does the term “LAYER” fit in?
A: The use of color in SpectreLayer’s graphics is simply a means to add appealing visuals. Although, the traditional definition of a spectrum represents light as 7 colors; -The term SPECTRELAYER has 11 letters. The simple act of coloring each letter with a unique color will inevitably create a band of colors (that neither conforms in color accuracy nor luminosity to a classical white line spectrum). But, it does look nicer than monochrome. Generically, there are many tools that analyze audio by frequency & provide processing based on analytics; -But Spectrelayer uses multiple analytics across the various aspects of the sound-field. AI utilizing tensors produce additional layers of data that must be processed. Some may be linear, non-linear, etc. The colors are not the primary focus of the Spectrelayer term any more than the color of a sports-car has bearing on its performance or what is under the hood. All of this audio data (tensor & otherwise) constitutes the “layers” of information processed during Spectrelayer utilization. The computers use these layers of data to provide processing guides & parameters for working towards improvement solutions. Layers of data (often represented by color keys) are processed by the “ghost in the machine” to create a “beautiful audible spectre”: SPECTRELAYER! BTW: Ogres also have layers – or so it’s been rumored
Q: Is Spectrelayer like one of those services where you upload a project & computers do the final mastering?
A: No. Spectrelayer is completely different. It is NOT a “music-mill”. As a matter of fact – you probably should consider using the Spectrelayer service AFTER you use any of those other services. Make the sound as good as you can and then “Spectrelayer it”. To find out why – take advantage of the Proof-Positive FREE Offer made below. You’ll hear why. Audio is a gigantic industry. There are lots of awesome tools, services & ventures designed to help the artist along in the many phases of production. Without those, there would be no niche for Spectrelayer – so, there is great gratitude that so many talented people are there to offer assistance. What Spectrelayer does is very specific & to our knowledge – there is no other similar service. Spectrelayer requires a competent technician. It is not a process where a file goes in one side & minutes later comes out of the other all finished (Wow! That would be a miracle)! Computers do the work but a tech MUST steer the process based on the source material’s unique sound profile. Use as many of the “others” to make your project awesome. And then use Spectrelayer as the LAST step in the production chain! A FREE TEST (described below) can easily demonstrate why. After all: Why would you trust a sales pitch more than your own ears!
Q: Can I use the Spectrelayer service to do the final mastering as opposed to post-mastering? A: Yes - some do. Simply send your final mix as a lossless stereo track in 24-32 bits in depth or as a complete project formatted for a DAW such as Reaper. The Spectrelayer service will treat it the same way as any track submitted to improve it. Taking this approach may represent an effective short-cut for some studios. In any case, you'll be able to audition the final sound before paying anything - as usual.
Q: Does the Spectrelayer service use AI to sample audio from external sources for voice or instrument emulation? A: NO! Such a practice might be seen as a form of copyright infringement. Spectrelayer uses algorithmic processing - not sample substitution from machine learning!
Q: Why are the Spectrelayer’d FLAC files on the opening-page’s Demo section all larger than the original FLAC files?
A: Because the Spectrelayer’d versions contain more detail than the originals. You see, “FLAC” is a lossless compression schema for making audio files smaller than the sources they came from while keeping the bit-perfect original file data intact. The more redundant material there is in a file – the better it can be compressed. Better compression means a smaller file. Greater detail is less compressible yielding a larger file. The fact Spectrelayer’s FLAC files are larger than their FLAC’d sources is additional proof that Spectrelayer files contain more audio detail (12-25% more on average)!
A: It’s bad. Some of the biggest purveyors of proverbial “snake-oil” in audio circles are well polished & well funded in the background. One of the latest scandals has to do with a process called MQA. A number of independent audio enthusiasts have called it a multimillion dollar scam and the evidence is piling up against it. Of course – audio has always seemingly been plagued with scams that seem to take advantage of the over-inflated egos of certain people who claim to have ears so golden that they can hear the fine ruffles of The Emperor’s New Clothes. These scams range from nonsense devices that are designed to shave the edges from compact discs to stratospherically-priced cables to endless variations of “Hi-Res” Audio schemes that all purport to use more bits & higher sampling rates than the CD Standard of 16bits/44.1kHz per channel (as if that standard is somehow antiquated or lacking). If you lay aside the promoting testimonies of people who have personalities likely defined by the DSM; – ALL of these scams have (1) commonality: The average person simply can’t hear the alleged improvement/s that are extolled by the promoters of these gimmicks or by the soft-personalities of those people whom the placebo effect also works wonders. If you need to strain to hear the improvement than it’s not real. Furthermore – test equipment can provide evidence that such claims are false. Spectrelayer, on the other hand is a process where the differences between source & output virtually jump out at the listener. Analysis of the same shows clear differences in the signals that are much more complex than mere loudness variances! Most people’s jaws drop open when hearing the Spectrelayer’d version of audio contrasted to its source. If you haven’t yet taken the time to audition the examples on the opening page – now would be a good time. Trust your ears!
A: Yes. If the process cannot remarkably improve the sound then the client should not consume time & resources. I.E: This may occur with a fantastically mastered classical or jazz recording. It would be rare but not unforeseen as a possibility. The second type of job that would be turned down would be a recording that had garbage for lyrics. Generally – music that celebrates the destruction of people, property or principles would get rejected. Ethics matter. We reserve the right to turn down such material or withhold the use of our logo on material that would damage our social or professional image.
A: Spectrelayer restores harmonics that have been lost or diminished, & reconstructs transients. Both often have high-frequency signal components that look like signal spikes (See scope image below). 16 bits is obviously the coding limit. That’s what’s recorded. However, of the people who have high-end D–>A converters that leverage oversampling – an interesting side effect often occurs. This involves the math generated by the oversampling process that interpolates & rebuilds the waveform transients between the actual data samples of a high-frequency peak – digitally. Such converters can create digital data that exceeds 16 bits in value based on the interpolated shape of the wave that was being described by the real adjacent peak samples in the 16 bit domain! And, if those converters have adequate voltage-supplies – then they can create signal voltages that will steer towards these higher transients! The next thing you’ll notice is that your power-amplifier is producing, I.E: 200+watt momentary transients instead of the 100-watts that a lesser D–>A converter would have demanded. Your tweeters will know it’s there. But will you? Only the best hearing can detect the differences. Perhaps your’s!
A: Spectrelayer shines on streaming platforms – louder and clearer than the loudest recorded traditional music! This fact almost seems counterintuitive because Spectrelayer’d tracks are not as loud as their source originals. The extreme irony is that streaming-music services look at the overall loudness of each track and reduce the volume of that track in order to maintain a median loudness profile. Because Spectrelayer’d tracks are less loud on average – they are penalized less (sometimes not at all)! And, because Spectrelayer’d audio has an improved dynamic range – all of those transients that make the music sound better make it through the streaming loudness algorithm unscathed! I.E: A “super-loud” studio release that averages -8dB LUFS would get attenuated about 6dB with most streaming services to achieve the industry target volume of -14dB LUFS. The SAME material in Spectrelayer’d form may realistically average -14dB LUFS as-is. This means the streaming service would not attenuate that Spectrelayer’d track at all – and the entirety of the dynamic range would be streamed! The result would be a fantastically better sound – completely unimpaired by the streaming service! Compared to Spectrelayer in that context – the original would sound pale, flat & lifeless. A quick note: The examples on the opening page of this website (original & Spectrelayer’d) sound very much like they would on a streaming service because those levels have been adjusted to approximate for loudness.
Q: The demo files on the Overview page seem to have varying sonic qualities among the different Spectrelayer’d versions. Why is that?
A: 
There are (3) reasons. Spectrelayer’s settings can be set over a wide range of parameters during postprocessing. Those files attempt to showcase various processing combinations. Each studio that produced the source recordings made each album with a sonic signature resulting from the mastering pipeline each was forged within. In addition, some of the source files are recorded better than others. While Spectrelayer can improve poorly mastered files dramatically, the general rule is that the better the source – the better the Spectrelayer’d result. I.E: The Linkin Park demo was from a severely damaged, loudness-war casualty. The Spectrelayer’d version is much better but the severe compression & distortion in parts of the original didn’t leave much good data to work with – as the original studio release was so damaged. However, the up-side is that tracks damaged like that provide a good source of training-test data for an AI. Several improvements in Spectrelayer since initially processing that track have yielded noticeably better results. The Phil Collins demo original (on the other hand) is so well recorded that it typifies the excellence possible with conventional mastering. Where Spectrelayer shines in post-processing such a file is the manipulation of a number of sonic parameters that the brain cues upon. No human engineer can hear nor track all of those parameters for optimization during the original mastering. Spectrelayer can; -& effectively rebuilds the waveform – leveraging what the system “knows” sounds good to the human brain. Same track; – Different emphasis on the sonic spectrum & dynamics. Early on it dawned on me that there are components to sound that human-hearing may not be adept at detecting on a conscious level. This is true about light & the perception of color so why should we think we hear in the best possible schema, either? All that rumination got me thinking: “If we could find a way to process subtle properties of the audio as certain creatures like the mantis-shrimp process unique colors (perhaps with a specialized AI); -THEN maybe a signal could be reconstituted combining traditional with the newer processing schemas that sounds better than the sum of the parts? And most people agree: Spectrelayer often sounds substantially better than the source audio & would describe the sound of Spectrelayer as “Je ne sais’quoi“.
Q: The Spectrelayer’d versions seem to have a slight tonality bump on the extremely low & high ends? Why is that?
A: There may be several reasons you are hearing unanticipated level differences on the spectrum ends. The most obvious is the fact that Spectrelayer is designed to be played back LOUDLY. However, I’m not talking about levels that damage human hearing. See, high fidelity reproduction also implies proper level playback that approaches “live” sound. At low volumes, the human ear is simply less sensitive to the ends of the audio spectrum. This is why so many early preamplifiers were equipped with a “loudness” switch. With that switch engaged, the tonal curve received a variable bump-up on both ends – the lower the master volume was set. The existence of the “loudness” setting was a tribute to engineers who knew that this effect was very real. Omitting it over time was a step backwards & purely as a cost-cutting measure (regardless of the excuses trying to justify it). Without tonal compensation at low levels – music sounds “thin”. However, as the playback volume approaches realistic high-fidelity levels, the Spectrelayer tonality curve shines. This volume is fairly easy to achieve with headphones (one of the most popular playback transducers used today). Home audio requires a bit more power. Generally, audio systems bearing the THX logo are quite Spectrelayer friendly & up to the task. However, depending on your hardware, you may find that some tonality compensation is still in order. Why?
Rather than pretend that we live in a perfect world, -Spectrelayer acknowledges that we don’t. As a result, Spectrelayer uses statistics to create a tonal curve that addresses the most common limitations of audio hardware & human hearing in advance; – & has made it possible for the end listener to achieve the best sound for their unique system by mildly tweaking only (2) tone controls: The Bass & the Treble. The primary reason that some people may need to adjust the Bass & Treble is that the extreme ends of the audio spectrum are where most weaker-tier components reveal their shortfalls. Many loudspeakers roll off the high frequencies abnormally – dulling the upper harmonics (hello Bose). The same is true about the Bass end of the audio spectrum. So Spectrelayer creates balanced, dynamic tonality curves that anticipate human hearing aberrations & playback hardware shortfalls. Because Spectrelayer preserves bit-depth to reproduce nuances – there is often audio data available to describe the subtleties in the dynamics of the low & high end that the ear can indeed resolve if the recording contains these cues. And, since it’s better to cut too little of a tonal range than too much; -Spectrelayer is comprehensive regarding these factors & takes steps to work these detailed nuances into the mix by adjusting the overall tonal distribution on the audio spectrum. This approach has (2) effects: People hear highs & lows with better detail & clarity than ever before in recorded material; -& audio systems that may normally suffer from internal limits – sound better than usual. If your sound system already resolves the ends of the spectrum with great clarity then simply adjusting the respective tone controls will provide the level adjustments (up or down) that you need to best enjoy the Spectrelayer sonic signature. How Spectrelayer deals with these common shortfalls is complex. Adjusting for your unique situation after the fact – is much simpler. Once made, the adjustment tends to work well on most Spectrelayer’d content.
A: Yes. Discounts for Exceptional & Humanitarian Project Ventures … I.E: “We Are the World” genres. If you are involved in a project that has humanitarian merit such as projects that support charities, or are focused on the betterment of humanity via altruism (even if you are merely encouraging the actions of such in lyric); -a special discount rate structure may be available to process your audio via Spectrelayer. Please reach out via e-mail & explain fully what you are doing. The savings could be substantial. This is especially true for bands & art-centric organizations that encourage charitable & responsible life choices. Spectrelayer would like to help you shine! Savings of up to 80% are available as are “Pay Us Later” opportunities based on the merit of each track processed.
A: Every situation depends on a large number of factors. Briefly – it depends on the quality of the source file. As a rule-of-thumb: Sow’s-ears are occasionally made into silk-purses; -But ckicken-sh!t can never be made into chicken-soup.
A: Yes. Generally speaking – the client can request any common lossless format. However, keep in mind that – the chain is only as strong as its weakest link. See previous comment.
Q: It has been rumored that the Spectrelayer schema was developed -in part- to leverage the Cannabidiol systems of the brain. Is Spectrelayer electronic THC?
A: This rumor is predicated on a tid-bit of truth. However, the bigger picture is that Spectrelayer was designed to affect several centers of the brain involved in sound & music processing. And I say “Sound & Music” because, technically – they’re not exactly the same mechanism. You can imagine these centers of the brain each as a “layer” that the Spectrelayer schema applies strategy to. The most obvious is the amygdala. If you’ve ever been awakened by a loud noise that made your heart pound & made you feel like you had been drinking overly-strong espresso … that’s the amygdala. In less intense situations – the amygdala adds a sense of excited-appreciation – such as when you get surprised by a birthday party or benefit from unexpected luck. Certain sound patterns can trigger it repeatedly & Spectrelayer is attuned to doing so when possible. The next layer of mental processing Spectrelayer is adept at triggering is the sensation that comes with being able to clearly recognize & sort multiple sounds simultaneously. This processing is often amplified in many people while using marijuana recreationally. The question arose: “What’s happening in the brain while a person is experiencing this?”. To make a long story short, THC changes the propagation speed of synaptic firing – stretching out the period of time that virtual circuits (called strange attractors) exist in the brain. Higher brain centers therefore have more time to examine these signals before they degrade & this creates the sensation of hyper-perception. Because this sensation is often connected to music processing & the common experience is one of extreme pleasure while listening – the question that arose was: “Can a similar effect on the music-processing centers of the brain be done by restoring the sound field to a better approximation of real world audio – without THC?”. I believe the answer is “Yes” & Spectrelayer attempts this with an active equalization process combined with extremely tight dynamic range control. These “layers” of processing are why music encoded with Spectrelayer is designed to be played back loudly. It’s NOT about being continuously “loud”. That would be uncomfortable. What it’s about is having that volume control turned up so that the amplifier can deliver the rarer, moment to moment peaks (which are often 10-1000 times briefer & more powerful) needed to generate these auditory conditions. In summary: These strategies work synergistically with the other aspects of the Spectrelayer process & are designed to improve the listening experience – never degrade it. Also – because Spectrelayer works with the sound-processing part of the brain – it cannot emulate any other effect of THC. So, if your long-haired friend “Percy” asks: No, you cannot get stoned merely by listening to Spectrelayer’d content! However, rumor has it – that in those places where doing such is lawful – combining THC with Spectrelayer’d audio “…takes you to a place“. And that’s all I can say about that.
A:
So, it is true that Spectrelayer’d content tends to be 5-10 dB lower on average in perceived loudness than its “loudness-war” damaged counterparts. In order to restore the extreme detail that makes music musical, Spectrelayer reallocated several bits of data per sample to encode the finesse, spacial cues & clarity of the music. This restoration of detail is WHY Spectrelayer is quieter. You see, a digital audio data stream only has so much data by which it describes the moment to moment sound. Spectrelayer chose to allocate those bits for detail – instead of continuous, unnatural, loudness. However, seemingly unknown to many people in the recording industry & radio broadcasting; -There IS a high-tech device capable of making Spectrelayer’d content sound louder WITHOUT SACRIFICING DETAIL OR SOUND QUALITY. That’s right! A technology exists that allows Spectrelayer to eat its loudness cake & have it too! It also turns out that this technology has been around for quite some time. Perhaps it’s on your music device, too! On whatever device you’re using to play back your music, look for a controller with the letters nearby it’s location that spell the obscure term: “VOLUME“. This “VOLUME” controller may be a knob, slider of [^] like graphic. It turns out that adjusting this “VOLUME” control can make Spectrelayer content louder WITHOUT sacrificing the sonic clarity of the music! It’s my assertion that had the recording industry known about this “VOLUME” controller that much of the entire loudness war could have been prevented & many, many recordings that were damaged by engineers probably could have been preserved! If only the recording industry had been able to limit its overly-loud recordings to distribution to radio-stations, perhaps as special release versions (as if anything like that could occur)! Unfortunately, recordings damaged by the loudness war can’t be “fixed” by turning that amazing “VOLUME” knob in the other direction because, sadly, the musical detail was sacrificed in making the recording unnaturally “loud” to begin with. It’s been observed that: “You can’t UNburn a cake by freezing it.”. Fortunately, Spectrelayer can, proverbially, “recycle the old cake to make a better one using a brilliant baker named ‘AI”! 