mail us  |  mail this page

contact us
training  | 
tech stuff  | 

Survival Guide/Tech Stuff - Digital Audio/Video

Digital Audio or Audio/Visual is one of those areas that crosses the border between pure technical knowledge about computer'ish subjects - file formats, bit rates, sampling and all that good solid stuff, and other territory to do with sound frequencies and if the material is musical - tempo and pitch and harmonics and...and .... and the maths - oof! Fast Fourier Transforms (FFT) and RMS and Logarithms - stuff we thought we had long, and mercifully, left behind. We were quite shocked on our first excursion into the depths of this dark and frequently secretive world. And the shock when we discovered that you had to pay for some specifications. And Patents. What a lot of patents. We have only recently recovered.

Health Warning: These pages are a work in progress and are intended to capture our (very) slowly growing knowledge. At this time we are mostly interested in digital audio. We cover video where we stumble accross something related to audio. Our primary motivation is the preservation and restoration of old recorded music. If you spot an error, omission or incomprehensible description please take the time to write and point it out by using the links at the top or bottom of each page. We always respond in the same spirit in which you write to us.


Sound Primer Fairly quick overview of Frequencies, Notes and Pitch (even ADSR) together with all that Harmonics, Overtones, Fundamentals - and even Partials - stuff, followed by the mystery of Loudness and sound power. Necessary but horrible stuff, just horrible.
Digital Sound Primer Big page. Starts with an overview of analog sound and recording history (well, where would you put it). How audio digitization works, sampling theory (Nyquist), quantization, Time domains, frequency domains, looking at ADSR and harmonics. Lie down in a darkened room for a minimum of 3 hours after reading this.
Frequencies Frequencies of musical notes together with some information about typical instrument (and voice) frequency ranges and other incidental noises that you may come accross in your daily life.
Equalization & FFT Equalization, with Preferred and Calculated center frequencies (kinda explained). Sound metering (VU and Peak and all that other nonsense) with a description of what we did to code a useful (to us) sound meter. Finally, no discussion of equalization is complete without the good old FFT and boy, do we discuss the FFT. These are notes about our implementation - or 'trying to provide sensible output when they don't tell you stuff'. Headache inducing page, and big, very big. But useful, if you like this kind of thing.
Files & Codecs Quick overview of files, codecs and containers - all that alphabet soup of AAC, HE-AAC, AIFF, MP3, MPEG, m4a stuff. Pretty complete - well, the more inportant ones anyway - but not exhaustive.
Acoustic Calculators Note frequency to various decimal plates and user defined A4 pitch frequency. Preferred and Calculated (band) frequencies for a user defined range of octave values (1 to 1/48). FFT/DFT duration, center and band frequencies for a FFT/DFT size, sample size and bit-depth. Heady stuff. Not for the faint hearted. But jolly good fun.
Glossary Some stuff we picked up along the way. Just another one of the 10 million audio glossaries available on the web.
Audio on Windows 7 We had a lot - a lot - of trouble to get a glitch free audio playback on a Windows 7 laptop. Turns out we were not alone. This is how we fixed it. While the specific method may not work for you we focuss on the methodology we used which should work for everyone.
Audacity notes Terrific utility. Some things we learned along the way - not terribly interesting for anyone who is a regular user of this great tool.

Our History, Needs and Plans for Digital Audio/Visual

History: The Digital Audio project is partly professional, partly hobby. We started trying to restore a number of digital audio tracks that had mostly been captured from very old 78 RPM records, occasionally vinyl, but mostly good old 78's (shellac) in the majority of cases from the 1920's, 30's and 40's. In some cases we had access to the source material and could re-capture but mostly not. We had what we had and were expected to perform miracles. Which, as we all know, can take a bit of time. Professionally we are increasingly involved with some serious IPTV work which is bringing more visual stuff into the picture and which will, over time, be increasingly reflected in the content of this section.

We started to use Audacity and that took us a long way. The noise removal and audio repair functions are very impressive. Then we moved into equalization in order to try and improve sound quality. We made some bad mistakes here. But since we always kept the original files nothing catastrophic occurred - we could always start again. But there are limits to Audacity when using some services - that does not detract in any way from the excellent quality and functionality for many, many other purposes and we still regularly use it. In particular we found that the batch mode of spectrum analysis extremely limiting. Of necessity it uses an averaging method to compute the spectrum graph and initially this was sufficient.

Then we got greedy. We wanted to be able to see the dynamic frequency. So we experimnted with a number of real time frequency analysers. Then we got very unhappy - some used 9 bands, 7 bands, 12 bands. Some use green and red leds to show intensity (what's intensity and does that relate to loudness). Almost none provided any useful documentation of their selected frequency bands or the reasons for the colors. Some came with equalizers which almost never were tied to their frequncy displays. A 10 band frequency analyzer with a three position equalizer (bass, middle and treble) is not too exciting especially when it does not even define the cut-offs for each range. Seemed this was more of an art form than a serious endeavour - not that artists are not serious people. Some of them are extremely serious.

If we were prepared to pay $gazillions we could get all we wanted. But we were not. So we finally found a reasonably spec'd Open Source player with what looked like a decent equalizer. So using our ears as a frequency analyzer (not too reliable) we played with that and it took us further down the road.

Finally we got past greedy - we got gluttonous. What we really wanted was:

  1. A real-time frequency analyzer with at least 1/3 octave (31 bands).
  2. The ability to drill-down even further within one or more 1/3 octaves - 1/12 octave or even lower
  3. VU-Meter/PPM with peak detection showing on a dB (SPL) scale not these meaningless undimensioned scales of the original meters. These was an excuse 70 years ago when this stuff was originally developed but not today.
  4. An equalizer that would reflect directly the frequency analyzer (1/3 or lower). The idea being that we could set the equalizer directly on top of the frequency analyzer display and fool around with it in real-time.
  5. A hard limiter on the equalizer, so we could set the gain level but be able to cap it for any given frequency or accross all frequencies.
  6. The ability to save the equalizer settings as a profile that could be recalled in various ways including merging multiple equalization profiles.
  7. An equalizer that would be optionally harmonic sensitive - we could construct a profile so that if we were picking up (amplifying),say, a violin we could define a harmonic profile that would automatically apply proportional gain to the first, second etc. harmonics. Eventually we could build a series of these and call up one or more, for example, boost the violins by, say, 6 dB(SPL) and reduce the violas by, say, 4 dB(SPL). In real-time.
  8. We wanted to be able to change equalizer settings at various times within a track. Thus the ability to have multiple equalizer profiles that we could set against the track time. Use equalizer profile 1 until 35 seconds, then switch to profile 7 until 1 minute 47 seconds etc., etc..
  9. Since in our old recordings we felt that we had some rogue frequencies we wanted an equalizer that could move through a selected range (scan) and knock out (or boost) one frequency range at a time (with or without harmonics) to see the effect.
  10. Since much of the material we were working with had, at best, a dubious provenance and had probably been disatrously mangled - possibly even by us in an earlier life we needed the ability to 'clean' such material. While term 'enhancer' is widely used to describe multiple capabilities we wanted the ability to restore the instrument characteristics. In short, given an intrument's starting point (not always a simple thing to identify) the 'enhancer' would add or adjust harmonic frequencies based on that instrument's characteristics - including its ADSR profile.
  11. Clearly we wanted to able to both playback the effects in real-time and/or capture to a file in at least WAV and FLAC format (we could subsequently convert to any other format required) once we were happy.

None of this seems too, too unreasonable and some of it is pure speculation on our part that it will yield the reults we want. At this time we have decided that we need to develop our own tool-set. We like Open Source but will release our code under the more permissive BSD/MIT license rather than GNU GPL. These pages are the result of our current development work and our pathetic, plodding and unbearably slow aquisition of knowledge (well that's how we like to characterize it) in audio and visual processing.

Don't hold your breath waiting for this stuff to appear.

Problems, comments, suggestions, corrections (including broken links) or something to add? Please take the time from a busy life to 'mail us' (at top of screen), the webmaster (below) or info-support at zytrax. You will have a warm inner glow for the rest of the day.

Tech Stuff

If you are happy it's OK - but your browser is giving a less than optimal experience on our site. You could, at no charge, upgrade to a W3C standards compliant browser such as Firefox




share page via facebook tweet this page


email us Send to a friend feature print this page Decrease font size Increase font size Display full width page



Theory Stuff

Filters, FFT and more
Signal Processing
Audio glossary
Numeric Algorithms
UNSW Acoustics Info
UNSW - Dreaded Decibel
Acoustic Explanations
FFT Primer and Book
FFT Output

Interesting Stuff

MP3/OGG Stuff
Instrument Samples

Our Stuff

Intro and Overview
A/V Formats
Audio Frequencies
Sound Primer
Digital Sound
Equalization, Meters, FFT
Acoustic Calculator
Audio Meta Data
Win7 Audio Problems
Audio Glossary


CSS Technology SPF Record Conformant Domain
Copyright © 1994 - 2017 ZyTrax, Inc.
All rights reserved. Legal and Privacy
site by zytrax
Hosted by
web-master at zytrax
Page modified: October 21 2015.