Droid Does!

Yes, your Android phone can put out a wicked sounding Transformers style voice.  At least, with a little help from the Caustic Digital Audio Workstation.  In this article, I'll be going over the new Vocoder synthesizer in the upcoming release of Caustic 3.

According to David Battino, Vocoders have been a novelty staple in pop music for over 30 years [1]. You have most likely heard early Bon Jovi songs where these have been used, and especially Pink Floyd songs.  Vocoders are big in many electronic music genres as well.

The Vocoder takes a sample, usually of a person's voice, in standard .WAV file format.  Then, you play the sound back using the keys on Caustic's keyboard, or in piano roll view.  Next, you tweak some settings to get the synthetic voice you are looking for.  Key thing to remember is that the lowest 6 notes of the keyboard in each Vocoder machine are assigned to the samples you load up in the Vocoder (each Vocoder machine can hold 6 different samples).  Then you need to add a signal note on the key you want your sound to be played on.

That last bit is important because if your sample was performed at a particular pitch, you'll want to assign the sample to that same key on the keyboard.  You can figure out the root key of any sample using the sample editor in the Vocoder.

 Let's walk through how to use this synth, so grab your Android device, start Caustic, and read on.

Adding a Vocoder Synth

Welcome to Caustic 3

When you start Caustic 3, you'll probably see the same screen as shown above.  To get started with this Vocoder tutorial, touch the 'Machines' button:

Caustic Machines Button

Then press the 'plus' symbol, in any of the spaces marked '1-14', and select a VOCODER from the screen shown below:


Now, touch the Vocoder instrument in the machine list view to jump to the synth's control screen:

Vocoder Added to Machine List

Now you have a Vocoder added, and you are ready to start creating some synthesized voices.

Default Vocoder Screen View

Obtaining a Speech Sample

I'm going to assume that you want to use the Vocoder with a sound file containing a speech sample.  Be just to get your gears turning, just because the Vocoder was designed around the idea of synthesizing a speech sample, doesn't mean that you always have to use it that way.  You could use it with any sound file you want.  Who knows what you might come up with?

There's two ways to get speech into Caustic really, and that's either by using your Android's built-in microphone to record a sample, or by loading a .WAV file containing speech from another source, outside of Caustic.  There is no 'correct' way; you need to pick whichever way works best for your goals.

If you want to use the on-board microphone, then you need to press the waveform button:

Caustic Vocoder Waveform Button

and then press the 'RECORD' button:  

Caustic Record Button

You'll see a window pop-up similar to this one:

Caustic Recording Input Source Window

I say 'similar' because I don't have a microphone setup on my machine, and you will most likely have one on your device.  Anyways, hit the big 'RECORD' button, and speak into your microphone, then hit the stop button to end recording.  You should see your recorded sound on screen as a waveform.

Caustic Vocoder Recorded Sample

The first thing you need to do with this sample is crop it.  That means you define the actual start and end points of the waveform.  Otherwise, when you play it, you may have an unwanted delay at the beginning of the sample, or additional dead space at the end of the sample.  We just want to hear the main sample, so drag the dark gray triangles in the waveform display to define the start & end times that you want.

Cropping a Sample in Caustic

When you have defined the start and end points you want, go to the 'EDIT' menu, and touch the word 'CROP'.  If you don't like it and want to redo it, then now is your chance to hit the UNDO command in the same menu.

The next thing you'll want to do is to 'NORMALIZE' the sample, which means that the volume is increased to be louder, but not distorted.  If you don't do this, you'll end up with a noise similar to 'tape hiss' (you remember cassette tapes, right?)

Touch the 'PROCESS' button, and Normalize is right at the top of the menu on the left.  You can also specify which way to normalize:  louder, or softer.  To do this, touch the button next to the word normalize.  Push the slider higher to normalize louder, or lower to normalize quieter.

Caustic Vocoder Sample Normalize

When you are done normalizing, you can check what the root key of the sample should be by touching the 'TOOLS' button, and pressing the button 'FIND ROOT'.

Now return to the 'FILE' menu and touch the 'SAVE AS' button, and then give your sample a name, and save it.  Touch the 'DONE' button in the lower-right corner of the screen after you have saved the sample.

Processing the Speech Sample

Caustic Vocoder Sample Loaded

Now that you have a speech sample loaded, you can start giving it character.  What better place to do that than at the Character Control sliders?!  These function similar to a graphic equalizer, by emphasizing or de-emphasizing the amount of sound heard at various frequency ranges.  With the sliders all in the middle, nothing is being made more or less noticeable.  Start moving the sliders around until you get a sound you like.  The effect will be present, but not overly obvious.

The rest is all about tweaking the dials to get the sound to sound the way you want it to.

Touch the Waveform selector to pick either a saw tooth or square wave.

The 'Unison' dial will make it sound more or less 'synthetic'.  

The 'Sub' dial will change how 'narrow' or 'fat' the frequency range sounds.  

The 'Noise' dial adds noise.  

The 'Slew' dial determines how much of your sample will actually make it to your ears, by affecting the selected waveform.  With the dial full left, pretty much all the sharp speech sounds drop away, and with the dial full-right, everything is enunciated 100%.

The 'HF BYPASS' dial controls how much of the high frequency portion of the sample gets through to your ears without as much signal modification.

The 'DRY' dial controls how much 'un-adulterated' sound is present at the output, along with the modified sound.

Adding the Sounds to Your Music

All you need to do now is create a pattern in the Vocoder's piano roll view, and add the pattern to your sequencer.  Only there's 1 trick you need to know in order for it to work right.  Vocoders need a carrier note, and a signal note.  The bottom 6 notes of the piano roll are assigned to your samples, so you will need to place a note there.  You'll also need to place a note at the same time in the piano roll view on the key you want your sound to be played at.  Reference below:

Vocoder Signal and Carrier Notes

Hope you enjoy using Caustic 3's new Vocoder!  Looking forward to hearing your creations!