What Is The Difference Between Voice Synthesizer And Auto Tune

What Is The Difference Between Voice Synthesizer And Auto Tune Free
What Is The Difference Between Voice Synthesizer And Auto Tuner

Pitch Correction Plug-In [Mac/PC]

Well, a talk box is basically an effect that's applied to an instrument sound influenced (and controlled) essentially by the sound of your voice, actually using a physical tube. You can hear it throughout Bon Jovi - Livin' On A Prayer (e.g. The effect is created in software by setting it up so that it corrects pitch errors in an unnaturally precise and fast manner, creating the metallic gliding between notes. This effect isn’t exclusive to Antares Auto-Tune and can also be replicated in programs such as Melodyne, a more sophisticated vocal correction program. Feb 25, 2015 Quick look at the difference between autotune, vocoders, talkboxs and voice altering software. The programs used are Ableton Live, AutoTune, Vokator, Melodyn.

Signal Processors >Pitch-shifters

Auto-Tune has been the industry‑standard pitch‑correction tool ever since its 1997 release. Does the latest Evo version keep the competition at bay?

Graphical editing now includes note objects as well as curves, and object-specific settings for the Retune speed.

Love it or loathe it, pitch‑correction software is now a ubiquitous part of the modern computer‑based recording studio. Antares were pioneers of this technology and Auto‑Tune became a clear industry standard soon after its release in 1997. Of course, other companies were not going to let Antares have things all their own way, and alternatives from the likes of TC Electronic/TC Helicon and Celemony have certainly made their mark. SOS reviewed Auto‑Tune 5 in a head‑to‑head with Celemony's plug‑in version of Melodyne in the March 2007 issue (/sos/mar07/articles/at5vsmelodyne.htm), and Antares are now back with what is, in essence, Auto‑Tune 6. However, the new version is named Auto‑Tune Evo, and includes some significant new features.

All About Evo

Auto‑Tune Evo retains the basic dual‑mode operation of earlier releases. An 'auto' mode provides easy‑to‑use, real‑time correction options and, if the original performance is pretty good in the first place, can often do the job with the minimum of fuss. If the tuning problems are more significant, the 'graphical' mode provides tools for editing indvidual notes via curves and lines.

All these functions are retained in Evo, but they have been accompanied by some significant new features and a reorganisation of the user interface. The most obvious visual change has been the expansion of the common controls section along the top of the main window. This includes the return of the Tracking control: in Auto‑Tune 5 this had been relegated to the Options screen, a move which apparently proved unpopular with many users. In the native versions of the plug‑in, this common area now also includes formant and throat‑modelling technology derived from the Avox 2 plug‑in bundle (reviewed in the October 2008 issue). The controls are basic — in simple terms, allowing the user to adjust the resonant frequencies and throat length of the singer's body — but used sparingly, they can produce a subtle shift in the character of the voice.

Real‑time pitch‑shifting is also included within the common controls area and is available in all versions of Evo. In most other regards, the operation of auto mode remains as before, so existing Auto‑Tune users should find the transition a relatively painless one.

However, the changes are not merely cosmetic, and the underlying pitch correction engine has undergone a complete overhaul, to improve the end results and to take advantage of the increasingly powerful processing available in modern personal computers. The differences in the engine are significant enough that Auto‑Tune 5 and Auto‑Tune Evo are not interchangeable, so existing users will need to run Auto‑Tune 5 alongside Evo if revisiting older projects. Potential purchasers should also note that pricing of new and upgrade versions of Evo vary depending upon the format required.

Tuning & Retuning

The Option dialogue allows the size of the Evo window to be adjusted and keyboard controls set.Electra vst torrent download.

The most eye‑catching new features are within Evo's graphical mode, where two additions really stand out. First, as well as the existing curve and line editing functions, graphical mode now includes 'notes'. At first glance, one can't help but be reminded of the way Celemony's flagship product operates. However, while Evo's notes do share some of what Melodyne's 'blobs' offer, as outlined below, the detailed functionality is different.

The second key feature is the provision of object‑specific Retune speeds. This might not sound like such a big deal but, in terms of making the graphical mode easier to use, it is a significant change. The Retune setting controls how quickly pitch is pulled back into line with the target note. In previous versions, this was a global setting — easy to use, but not very accommodating when you had one phrase that required a fast Retune speed but another where a fast setting produced an unnatural correction. The Retune control could, of course, be automated in the host sequencer, but in Evo you can select sections of the performance (for example, a section that represents a single word or phrase) using the I‑beam tool and then apply a unique Retune speed to just that selection. This can almost be thought of as a halfway house between auto mode and the time‑consuming curve editing of graphical mode.

The 'note' objects now provide an additional option for graphical editing. Once the Track Pitch function has been used to capture the actual pitch of a performance (displayed as a red curve), clicking the Import Auto button will produce a green curve that shows how the pitch is being corrected by the settings from the auto mode window. However, if the Make Notes button is pressed, Evo generates note objects and a green curve that shows the pitch correction within notes and the links between them. Note generation can be done for a whole performance or just a selected time range, and the user can control how many notes are generated. If engaged, the Snap To Note button forces notes to centre their pitch on the nearest 'correct' note. The useful Show Lanes button can also toggle on and off a shaded lane display for each note (similar to that seen in most MIDI note editors) and this can make moving note objects to different pitches a little easier.

Once generated, notes can be adjusted in both length and pitch; as they are moved, the green pitch curve is also adjusted. This mode lacks the very detailed editing provided by manipulating the curves themselves or drawing your own pitch lines or curves, but in most cases it is a very much faster way of doing the bulk of the editing required. Incidentally, only one graphical editing mode is available at any one point along the timeline — curve, line or note — but they can be freely mixed and matched for different phrases within the performance. The new 'note' feature is a big plus in terms of ease of use and, while the options for manipulating pitch within and between notes are perhaps not as flexible as those found in Melodyne, it certainly takes a lot of the work out of graphical-mode editing.

We Are Evo

I tested the VST version of Auto‑Tune Evo with Cubase. It performed very well and certainly didn't seem noticeably more taxing on the host system than Auto‑Tune 5. The only technical issue I encountered was that some of the graphical-mode tools occasionally went AWOL, although this was simply resolved by engaging the 'Plug‑in Editors Always On Top' settings from within the Cubase Preferences / VST / Plug‑ins options.

Running Auto‑Tune 5 and Evo side by side suggested that any differences in the quality of the pitch correction are subtle when using auto mode, although Antares suggest that Evo's new engine does make a better job of poorer‑quality signals. For me, the real difference was in graphical mode, where the variable Retune speed and note editing made getting a natural result with more problematic material considerably faster.

For serious Auto‑Tune users (that is, those that regularly get beyond auto mode), I think the new graphical mode features will be very welcome, and make upgrading well worthwhile. In this regard, Evo is a significant step forward from Auto‑Tune 5 in terms of ease of use, regardless of any more subtle (to my ears at least) improvements in the quality of the pitch‑correction algorithm.

Good though Evo is, I suspect the advances will not be enough to persuade most existing Melodyne users to jump ship. However, for those looking to buy into the world of top‑of‑the‑range pitch correction for the first time, the choice between Evo and Melodyne is a difficult one unless you can spend some time with both products. Both are excellent and, for 90 percent of pitch‑correction tasks, I suspect either product will get the job done with a minimum of fuss. For my money, Melodyne still has a slight edge when it comes to more problematic pitch issues or creative re‑pitching of a melody line. However, with the new Evo note-editing mode, Antares have significantly closed the gap and, if your vocal is already decent, Evo's automatic mode remains the most straightforward means of tightening overall pitch.

Alternatives

The most obvious alternative to Auto-Tune Evo is Celemony's Melodyne, with the plug-in version now proving to be most popular. It lacks Evo's auto-mode ease of use, but its note-based graphical editing is both slick and very sophisticated. For Powercore users, the other obvious possibility is TC-Helicon's Intonator HS, which was reviewed back in the November 2004 issue, and the same company also have pitch correction built into a number of their hardware products.

Pros

Graphical mode editing is much improved, with object-specific Retune speed settings and note objects.
A range of welcome improvements have been made to the user interface.

Cons

None, other than it still doesn't make my singing sound like Robert Plant.

Summary

Evo represents a very useful evolution of the Auto-Tune line, and is an obvious upgrade for existing power users.

information

Native version £299.99; TDM version £499.99. Prices include VAT.

Sonic8 +44 (0)8701 657456.

Native version $399; TDM version $649.

Antares +1 831 461 7800.

Test Spec

Antares Auto-Tune Evo 6.0.7.
Intel quad-core 3GHz Q9650 with 4GB RAM running Vista 32-bit, with TC Electronic Konnekt 24D interface.
Tested with Steinberg Cubase 4.5.2.

Vocoders have been around since the method was used first to synthesize a human voice in the 1930s. The word itself is derived from combining the words VOice and enCODER. For a brief history of the Vocoder and some of the artists that have used them, see Ian Vargo’s excellent article here.

Auto-Tune vs. Talk Box vs. Vocoder vs. Cross Synthesis

There is some confusion regarding the difference between these processes which should be cleared up first:

Vocoding is a form of Cross Synthesis, which is nicely described here by Julius Smith:

Cross-synthesis is the technique of impressing the spectral envelope of one sound on the flattened spectrum of another. A typical example is to impress speech on various natural sounds, such as ‘talking wind.’

Let’s call the first signal the `modulating’ signal, and the other the ’carrier’ signal. Then the modulator may be a voice, and the carrier may be any spectrally rich sound such as wind, rain, creaking noises, flute, or other musical instrument sounds. Commercial [hardware] vocoders used as musical instruments consist of a keyboard synthesizer (for playing the carrier sounds) and a microphone for picking up the voice of the performer (to extract the modulation envelope). (source)

In other words, only the frequencies that are present in the modulating signal are retained in the carrier signal after the envelope is applied. In software, vocoders can usually function as inserts on an audio track, as an instrument on a virtual instrument track using a sidechained input from another audio track or a live external source, or in the case of devices like Vocal Synth 2, either as an audio insert effect or virtual instrument.

Random thought: What would happen if you Vocode a Vocoded signal?

Auto-Tune is designed to sort of quantize a vocal signal to the closest pitch or scale degree. You begin to get that robotic vocoder-esque sound when you push it hard so that the pitch jumps abruptly from note to note. But the process is significantly different from vocoding.

The Talk Box is also a very different process:

… is usually an effects pedal that sits on the floor and contains a speaker attached with an airtight connection to a plastic tube … the speaker is generally in the form of a compression driver, the sound-generating part of a horn loudspeaker with the horn replaced by the tube connection. The other end of the tube is taped to the side of a microphone, extending enough to direct the reproduced sound in or near the performer’s mouth. When activated, the sound from the amplifier is reproduced by the speaker in the talk box and directed through the tube into the performer’s mouth. The shape of the mouth filters the sound, with the modified sound being picked up by the microphone. The shape of the mouth changes the harmonic content of the sound in the same way it affects the harmonic content generated by the vocal folds when speaking. (source)

Carrier & Modulator

“One of the earliest vocoders was developed by Homer Dudley in the late 1920s/early 1930s at Bell Laboratories. Large-scale telephone networks were being implemented at the time, and the vocoder was developed as a potential solution to bandwidth issues. These early tools were designed to encode speech to reduce bandwidth, allowing multiple voice signals to be transmitted over the same cables. Vocoders even underpinned pivotal World War II technologies, such as the SIGSALY system used for secure voice transmission.” (Vocal Synth 2 Vocoder manual – iZotope)

(image source)

While the use of the vocoder is typically associated with the robotized voice you are undoubtedly familiar with, and the processing method can be used on any sound especially those with distinct transients such as percussive sounds. Don’t feel constrained by the name — try experimenting with sounds other than vocals — replace or parallel process rhythm parts for example to get a beat that tracks the harmony. I’m a big fan of misusing software for creative effect – unlike hardware; it cannot be broken with bold experimentation.

Below is a collection of software Vocoders (in no particular order) that I recommend based on personal use. I welcome any additions to the list from readers inclined to contribute. They all work great on vocals of course, but as mentioned above, I suggest a healthy dose of reckless abandon and misuse to get your creative juices flowing.

EVOC 20 PS by Apple

Price: $199 – included with Logic Pro X

This is the vocoder and polyphonic synth that comes with Logic Pro’s excellent suite of virtual instruments. (Check out my Synthesis 101 tutorial series that features Logic instruments here)

The interface is accessible and well organized — once the three modes (Analysis, Synth, Vocoder) are understood and the Side Chain input is set up, you will start getting immediate results. The Synth section (Carrier signal) can be a blend of two waveforms or an FM modulated waveform. Built-in modulation, noise generation, onboard effects and a cool Freeze function (which freezes the current vocoded sound to be used as a straight-ahead synth voice) are nice features. Voiced and Unvoiced detection is also available for increased intelligibility if desired.

Vocal Synth 2 by iZotope

Price: $199

This plugin by iZotope is extremely powerful and sounds great from the very first launch. It is addictive, so make sure you’ve got some extra time on your hands before you fire it up. There are three main modes to be aware of off the bat:

AUTO: The internal synth engine of the plugin is used to generate sound. Insert it on an audio track and press play.
MIDI: With this mode, you can control the harmonies generated with a MIDI controller (eg. Keyboard)
SIDECHAIN: In this mode, you can drive another synth, guitar or other sound sources through the Vocal Synth modules.

The help screens give clear and concise setup instructions for each mode.

In addition to the Vocoder Module, there are four others – Biovox, Compuvox, Talkbox and Polyvox. The interface uses a nice graphic to display the relative influence of each.

There is also a slew of audio effects that can be easily routed in any custom configuration simply by dragging. The Shred module, in particular, creates a great granular stuttering effect that’s worth the price of admission.

There’s A LOT of functionality in Vocal Synth 2 and I encourage you to dig deep.

MVocoder by Melda Production

Price: $57

The MVocoder is a feature-rich device that comes with an exhaustive 102-page manual and allows for the manipulation/modulation of every conceivable parameter. Four Modulator sections include morphable shapes, envelopes, followers and more. The Band Graph and Band Matrix sections alone make this a powerful tool. There is an onboard upsampling section and mid/side capability.

The more time I spent with it, the more I liked it as I continued to uncover new functionality, new graphic interfaces, and surprising layers of control that resulted in nuanced to extreme sonic variation. It even has 11 interface styles with custom colors to accommodate your aesthetic taste.

As with many plugins of this complexity, the learning curve is steep and to take full advantage of all that this device has to offer will require a significant time commitment. For what this thing does, the price is ridiculously low. But there is no price/functionality relationship here. The MVocoder is for those looking to explore vocoding in great depth with meticulous control and programmability at every turn.

Vocalizer Pro by Sonivox

Price: $199

The Vocalizer Pro has a stunning, albeit initially intimidating interface, and provides access to four identical spectral synthesis modules with customized routing capability. The modules include various synthesis methods, envelopes, multimode filters and LFOs. There is an Effects Tab for built-in EQ, Chorus, Delay and Reverb, and an excellent Browser system based on genre, color and character descriptors. Eight Snapshot pads allow on-the-fly access to user presets and eight Chord pads will recall keyboard notes and chords. This device sounds great right out of the virtual box and is almost immediately usable. Once you understand one module and the effect of the various synthesis methods you are ready to roll.

Lector Vocoder by Waldorf

Price: $179

The Lector Vocoder is a single window interface that includes four main sections: Input Carrier signal controls with Unvoiced detection for intelligibility, Synthesizer (2 oscillators, filter and amplifier sections), Vocoder filter bank with modulation and analysis controls, and an onboard Effects section that includes: overdrive, chorus, delay and reverb. It took some futzing to get this device to do what I wanted, but the results were reasonably good. I recommend a good read through the manual to avoid unnecessary frustration. The Waldorf sound engine is well-known and respected and provides a strong foundation for Vocoding in this plugin.

Orange Vocoder IV by Zynaptiq

Price: $189 (older version price)

The Orange Vocoder IV will be the latest offering by the innovative developers at Zynaptiq but has not yet been released, so I’m unable to make any comments or personal observations. However, check back after the release date for an update to this article. Until then, here is some preliminary information from Zynaptiq:

Initially developed by Prosoniq in 1998, ORANGE VOCODER has since evolved through numerous iterations, always staying at the cutting edge right to this day. It has a reputation for being the single most transparent sounding and flexible vocoder plug-in available … it features a built-in 32-voice synthesizer, eight different vocoder algorithms ranging from analog modeled circuits to far-out digital cross-synthesis modes, filter-bank freezing and reverb, a 10-band graphic EQ, and more…a dream-machine for creating new sounds!

I can say that the current version pictured below is a beautifully designed plugin that sounded immediately great from the very first instantiation. It features two sound sources which can be waveforms or samples, two modulators which be set to an LFO, envelope or dedicated pitch modulation, ring modulation, onboard reverb, filter, graphic EQ, unvoiced detection, mixer section, and a freeze button. I am anxious to see and hear what improvements the latest release will have to offer.

Conclusions

Vocoding is not just about voice mangling or robotizing; it is a unique form of cross synthesis that works well with a variety of source material and live audio input. If you haven’t experimented with vocoders in the past, now is as good a time as any. It may be just the thing to inspire new directions in your work.

Check out my other articles, reviews, interviews and my video tutorial series, Synthesis 101 available exclusively on The Pro Audio Files.

4reeaf4er.netlify.app