Hardware/Software Overview: The Speakerphone

A couple of weeks ago, I filed my first EU “On The Road” report by cycling across town to visit the Audioease world headquarters to check out the where, who, and how of things. Predictably, the piece was catnip to convolution reverb users and a lot of fun for me personally.

I was a little surprised to learn that a number of my Facebook friends who read the piece weren’t as aware of what else the Audioease guys do. Actually, I was maybe just a bit relieved, and maybe a little hopeful that they’d stay incurious. The truth is that Audioease’s Speakerphone plug-in (updated in version 2.1 to support 64-bit audio, dual platform, and available in Audio Unit, VST, and a host of other formats) has been one of my “secret sauces” as a recording artist for quite a while. At the risk of losing my mojo a little, let me see if I can tell you what I think makes this a pretty singular plug-in tool and maybe and talk about how it became my secret sauce as I go along (and hope I can find some new tricks so as to stay ahead of y'all).

Talking ‘bout A Convolution

The simplest – and maybe least reductive – way to talk about the place where the Altiverb and the Speakerphone diverge it to think just a bit about the process of convolution, along with a hint in the name of the software itself. The Audiease folks are probably best known for the Altiverb convolution reverb [Note: I am a devoted Altiverb user, and have been since its initial release. I am working to keep the convolution reverb fanboy to a minimum. I promise.].

Just in case you’re not familiar with what convolution is, it involves creating an “impulse response” – a picture of how a particular acoustic environment responds to frequencies across the frequency spectrum (using things like sinusoidal frequency sweeps or starter pistols, for example) and then transforming the resulting audio into a tool you can use and apply to any sound source you want. The Altiverb uses that technique along with a fantastic array of these impulse responses that the Audioease guys and their customers have created to make reverbs that sound like a certain place (and can be further altered). And by the way – Max for Live users have enjoyed some amazing Max-based convolution reverbs for quite a while now, along with Alex Harker’s HISSTools for MSP.

Life with the Altiverb goes like this: Once you finish a week or so of putting your banjo, MSP granulator patch, or that recording you got last winter of popcorn snow falling on your skylight left-front on the stage at the Sydney Opera House or burying it in a cave in Upstate New York or in a cooling tower you can step back a little. You'll begin to notice some impulse responses that have nothing at all to do with great concert halls or historically famous recording studio echo chambers. You can find and use impulse responses that let you put your signal in buckets or clothes dryers or big long plastic pipes. The technique is the same to work with any of those sources – and that’s where the the Altiverb and the Speakerphone begin to diverge in interesting ways.

Simply put, you can apply the source/filter convolution method to any linear system, which includes (hint hint) Speaker cabinets. The Speakerphone includes a dizzing array of ways to take your audio and run it through all kinds of vintage and modern amplifiers (I know people who bought this just to use it for guitar recordings), leslie cabinets, and far more exotic speakers. Megaphones. Old Tube Radios. Antique Victrolas. Telephones. Kid's toys.

From Supersession Amp to Secret Sauce

Perhaps the best way to summarize why it’s a go-to tool for me has its roots in the way I work with Max to create materials – I patch with the Speakerphone, and use it to generate and organize variety in much the same way I work with Max. The layout itself suggests the flow of processing – from modifying where the unprocessed audio is coming from (mic models) through a beautiful array of bit-crushers, limiters, delay lines, phasing/flanging/chorusing and – yes – Audioease convolution IR reverbs, too.

This fills in a whole bunch of the tools in my usual signal chain in a one-stop interface, throws in some cool effects beyond Parametric EQ, and – a joy for an LFO guy like myself to behold - a nice six pack of syncable LFOs you can click and drag to patch the rest of the Speakerphone controls.

One of the additions that it took a little time to get used to is what they call “Covers” – a way to obscure or muffle your sound by covering it with blankets, sticking it in the trunk of a car, or putting it in a partially opened wooden box. While it’s great for piling mufflers on your amp cabinet turned all the way up, things get really interesting when you start modulating the parameters that define the cover in real time using the MSP method of your choice.

And for Max users, we haven’t even started with the obvious and everyday possibilities offered for plug-in modulation using things like Max messages sent to the vst~ object that's hosting your Speakerphone. I suspect you’ll get around to this once you spend the first month just burrowing into the sonic riches right there on the Speakerphone’s front panel.

The interesting thing about using Speakerphone as a patching environment is that the whole process is available from a single front panel (don’t worry – you can fold it up to save screen space once you get things ticking along nicely) that’s elegantly designed.

In case you find all this a little daunting, never fear. The guys at Audioease have made a great introductory video that pretty much brings you to the place where I started fooling around and falling in love.

Fruit On The Bottom

If your only introduction and experience with the Speakerphone involves loading presets (and they are great, so that’s kind of understandable) – and even if you learn the art of the tweak-fu in Speakerphone, you may be used to thinking of those 5 columns at the bottom of the UI as holding area for the loops and one-shots that are used to “construct” various kinds of “scenes.”

Ever count the number of slots in each of the five columns? There are 12 of them. These are octaves. You’re looking at keyboard slots for a sampler - drag-and-drop slots for loops or one-shots. The samples used to create the standard Speakerphone presets are all available to load in one of the octave bins, ready for triggering using MIDI input. In addition to all those effects and convolution doodads, you've got a pretty nice sampler ready to roll.

A Note to Ableton Live Users: I had trouble getting my copy of the Speakerphone to receive MIDI with an Audio Units version of the plug-in loaded. I got in touch with Aram from Audioease about the problem, and was told that the VST version of Speakerphone is the version that supports MIDI input with Live. I’ve included a sample project that demonstrates how it’s done that you can download here to use as a starting point for your own investigations I’m grateful to Aram for his help.

• Download the Max for Live example files.

You can choose to include your sampler output pre- or post-effects with the click of the mouse. This lets you create interestingly balanced sampler and strongly modified audio input simultaneously and mix them on the fly (try a drum pattern composed of bad cable connection noises from the sampler as dry as a bottle of New Zealand red wine placed alongside a nice fat analog synth through one of their amp simulations placed in a room to put some “air” into the electronics, for example).

Okay, I haven’t given away every single Speakerphone trick I know, but that’s okay – this is such a well-designed (and CPU-friendly) piece of software that the combinatorial opportunities extend well beyond what I work with or am imagining for next week, even. While this software has been well received by lots of post-production houses interested in limiting the amount of special effects outsourcing they need to worry about, I think that the Speakerphone is also an environment for ordering noise in space in real time – it’s a composing environment par excellence.