1. Spectrogram Download
2. Spectrogram Waterfall Free Vst Plugin

Unlike typical 'waterfall' plots, Insight 2’s unique real-time 3D spectrogram creates a detailed topographical map of audio using iZotope's high-resolution spectrogram capabilities. Unfamiliar with a spectrogram? Watch our video to learn more. Choose between a 2D or 3D scrolling Spectrogram. Hi guys, Im just wondering if anyone knows of a cheap/free spectrograph vst that shows real time spectrum but also along the time axis. Something that looks like this. Audacity 2 size 8.52 MB Audacity is free, open source software for recording and editing sounds. It is available for Mac OS X, Microsoft Windows, GNU/Linux, and other operating systems.

1. BlueLab Wav3s is a plugin that simply displays the sound in 3D.Several display modes are available, as well as parameters to adjust the display. The view is interactive and can be easily manipulated using the mouse or the trackpad.
2. I love the spectrogram component that comes with foobar2000, it's interesting to be able to quickly and easily see a spectrogram of your currently playing music without having to start up an audio editor to access a spectrogram function. The customization options are great too, i just wonder if it's possible to add some more functions.

Encoding Images as Sound & Decoding via Spectrogram

by Gram Schmalz

Introduction to spectrograms and sonic bitmap encoding

Are you a circuit confident exploratory encoder, Aphex Twin fan, or a keen electro acoustic busybody? Have you ever wanted to turn images into sound, and then back into images? I think my answer to all of these questions was yes, and as such I have put many hours into figuring out the best, FREE, methods to do this. This article will give you a bit of a head start if you plan to do the same. Unfortunately since this article was written a few years ago, some of the links to examples and software are now broken. These have been crossed off, but left for reference sake.

A few years ago, I created an installation call Signal + Noise, which explored the splintered nature of contemporary experience through the visceral lived moment, and the digital document. A large part of this installation relied on spectrogram encoding and decoding. I enjoyed learning these technologies and thought I might share that learning experience through this tutorial.

here is one of the time-space documents created by the installation:

So, we want to make a sound image that is viewable on a spectrogram. But you may ask, what’s a spectrogram? Well, it’s quite simple. A spectrogram is “an intensity plot (usually on a log scale, such as dB) of the Short-Time Fourier Transform (STFT) magnitude. The STFT is simply a sequence of FFTs of windowed data segments, where the windows are usually allowed to overlap in time, typically by 25-50% [3]. It is an important representation of audio data because human hearing is based on a kind of real-time spectrogram encoded by the cochlea of the inner ear”[1]. In English, a spectrogram (also known as a spectral waterfall, or sonogram) is a time-frequency graph representing complex signals (such as audio) in an easy to interpret and analyze XY Cartesian grid format. On the graph: X (horizontal axis) represents time (depending on the spectrogram speed, the graph will move sideways along its X axis, showing the passage of time), Y (vertical axis) represents a frequency range as defined by the spectrogram. A third property (the one that makes it useful for image encoding) is also manifest on the spectrogram: frequency amplitude, or volume. In the image to the right, the red lines (spectrogram analysis of a whistle) are showing a high amplitude at those frequencies.

Now consider a bitmap image: a series of squares arranged in a grid with different values for each square. Within the spectral waterfall we have all the necessary elements to represent a bitmap image. The horizontal dimension of the image is manifest as time on the spectrogram, the vertical dimension by frequency (or pitch) and value is shown by the volume (or amplitude) of each frequency. If you think of the image as a series of frequency columns it is easy to understand how an image can be encoded in audio. Each column with its corresponding frequency amplitudes is encoded one after the other over time (representing X) to build up a representation of the photograph. Unfortunately using existing software it is impractical to encode colour images in this method, however black and white images are surprisingly easy to encode.

(if you are working on a specific project which requires colour images to be stored or transmitted in sonic formats, I recommend doing some research on Slow-scan television, a technique popularized by ham radio operators to transmit images back and forth. Ramon Glidden’s article “Getting Started With Slow Scan Television.”is a great jumping off point.)

Part two: the Jump off!

So you’re going to need some software. Fortunately everything you need is open source or freeware, unfortunately you’re going to need windows or linux, and so if you’re a mac user hopefully you have a second operating system installed on your computer. Once you get the software it is actually super easy, some of these programs can be slightly difficult to locate. I’ll include a list of links to all programs mentioned at the end of the article. First things first, you’ll need a spectrogram (or spectral-waterfall). There is a number of different open source applications out there for all different platforms, but some of them are slightly better suited to our use because they have more adjustable parameters. Here are my recommendations for this experiment:

At the time of this article being written, they are all free, so download the appropriate program for your use, and install it! Obviously I can’t give a tutorial on all these programs, or this article would be way to long, but the programs are all pretty self-explanatory. You just need to get the application running, and then find out what frequency range it is observing, so you can encode within that range.

If you need a little more info on how to use a spectrogram, Rob Hagiwara has written an extensive guide to understanding and reading spectrograms, which is available here:

Now you need an encoder, to change your image into sound. I have found three programs to use for this, two for windows, and one for Linux. Audiopaint and Coagula Light are the two Windows programs I located for completing this task (links at end). Both are free. Ohmpie (ohmpie.com) wrote a program for Linux to synthesize spectrogram images, and was also kind enough to share his code, and write a pretty good analysis of the mathematics involved in the process. I recommend checking it out even if you don’t use imageEncode, his Linux program.

A little summary of what these programs do: Each of these programs performs the exact same task we did when looking at the bitmap image earlier in this article. It breaks down the image into columns of pixels, and represents each column as a time interval. Then, looking at the Y-axis of each column, the program decides what frequencies (for pixel Y position) need to be played at what volume (representing brightness). The programs then use a multitude of sin waves synthesized at the corresponding frequencies and create an audio file that contains a simplistic rendition of the image.

Download the encoding program of your choice, and get it running. Once you have opened the program, you’ll need a bitmap image to encode with. Import one using the import setting, on all these programs, and have a look. You’ll need to change the audio parameters of the encoder, so look around and find the preferences panel that contains the Min and Max frequency parameters. You need to set these parameters to match the frequency range of the spectrogram you decide to use. You also need to look at the scale setting, there are a couple different scales you can use for these programs: Logarithmic, Exponential, or Linear. These pretty much just need to be set the same as your spectrogram. If the spectrogram you are using has an option for which one to use, Logarithmic seems to get the best image quality in my experience. each scale sounds different and they all function relatively well, so choose by how you want your images to sound. There is one other setting you’ll need to play with, duration, which controls how quickly the audio file plays, and also needs to correspond with the spectrogram program. Once you have the other stuff worked out, you can play around with this through trial and error and get it set right. The duration will also need to change for every different aspect ratio in the images you use. If the duration is set incorrectly, your image will be “squished” or elongated horizontally. You should be able to tell which one is which.

At this point you’re pretty much ready to encode and decode. Fire up your encoding program, import your picture, and generate your sound. Best modulation vst plugins. Fire up your spectrogram, and either import the file, or hook up a microphone or line input if you want to use the real-time method.

Hopefully this has been a useful point in the right direction. I know I didn’t reveal all the mysteries of spectrogram encoding, but where’s the fun in figuring it out if you already know all the answers? Doing some of the work yourself is where the fun and discovery happens, so get to it!

Coagula Light Tutorial: http://www.youtube.com/watch?v=dxJwsMrSaYI&feature=fvw

Spectrogram Download

Autumn – Image to Sound, by Binary Quandry (audiopaint example): http://www.youtube.com/watch?v=bya2qZChcvI

Spectrograms:

AudioXplorer (os x): http://www.arizona-software.ch/audioxplorer/

Sonic Visualizer (os x and windows): http://www.sonicvisualiser.org/

Spectrogram Waterfall Free Vst Plugin

Spectrogram (gram50) (windows): http://www.softlookup.com/download.asp?id=18945

Encoders (colour-note synthesizers):

Coagula Light (windows): http://hem.passagen.se/rasmuse/Coagula.htm

Audio Paint (windows): http://www.nicolasfournel.com/audiopaint.htm

imageEncode (Linux): http://www.ohmpie.com/imageEncode

[1] “Mathematics of the Discrete Fourier Transform (DFT), with Audio Applications — Second Edition”, by Julius O. Smith III, W3K Publishing, 2007, ISBN 978-0-9745607-4-8.