Friday, August 30, 2019

A brief history of Autotune


          By now, most people are familiar with the audio software application Autotune. Autotune is a pitch correction device used in vocal processing that was developed by Andy Hildebrand of Exxon, originally intended to manipulate sound waves for locating offshore oil reserves as well as for detecting earthquakes before they fully manifest. It was subsequently used by recording studios to correct vocal pitches on jingles, movies and other studio projects, and was a godsend for producers in session with artists whose material, persona and timbre of voice were enough to sell records, while their strict technical ability as vocalists was in some degree lacking. Techniques for achieving this same result previously existed in tape studios, but they were cumbersome and rarely used.
           The initial mission for Autotune was to keep the pitch adjustments as transparent as possible. Even when adjusting pitches in increments far less than a half-step, there are variations in formants and harmonic overtones that without compensation are noticeable to even an untrained ear and glaring to a professional or aficionado. Unsophisticated coding within software of this type can lend elevated pitches an “Alvin and the Chipmunks” kind of effect, and lowered pitches a mouthful-of-marbles kind of articulation. The true grace of Autotune’s coding was its sensitivity to the harmonic structure of pitches and the natural changes in formants that present themselves with pitch correction.
                The reasons for its immediate commercial success are obvious. Imagine an advertising executive booking a jingle singer and a producer for a TV commercial and neither of them are 100% that evening. The singer has already gone home with his check but a later playback reveals a pitch problem on one of the words. What used to be a disaster suddenly became easily manageable.
                Then came the inevitable experimentation and creative mangling of the initial purpose of the software by creative artists, as is the way with all new tools in music. Dance records in the late nineties used it on lead vocal tracks and it became a well-known club sound in England and the US. The first mass popularization of the unabashed use of Autotune on a lead vocal was Cher’s number one Billboard hit, Believe, in 1998.
                Listener or professional music-maker, everyone interested in sound all of a sudden asked themselves two questions: “What the hell was that?” and “Why am I listening to a Cher record?” Believe became Cher’s biggest hit ever, leaving Gypsies, Tramps and Thieves, Half-Breed and other racially tone-deaf anthems in its unwavering pitch and sales wake. Whatever its artistic merit, Believe featured a lead vocal sound millions of Americans had never heard before and they loved it. After the Cher phenomena, Autotune use, with its signature audible click as it releases one pitch and jumps to another, was soon ubiquitous on the radio. Artists emerged whose every vocal take was processed using Autotune, notably T Pain, who scored the ultimate pitch-correcting triumph in his use of Autotune often being referred to as “The T Pain Effect.”
                Autotune is by no means the only pitch correction game in town. Melodyne aims itself pointedly toward the market that serviced the initial use of Autotune, to correct pitches with utter transparency rather than to use it to obvious effect on a lead vocal. Melodyne introduced a radical component to its software suite a few years ago called Direct Note Access or DNA, which lets you isolate individual pitches within a polyphonic performance and manipulate its attributes without affecting other notes in the chord. That is to say, if you are recording a barbershop quartet on a single microphone and you get a fantastic performance, save for Uncle Floyd’s one flat note in the last verse, DNA can isolate old Floyd’s baritone clam and bring it into pitch. It’s on the other three to pull him aside and tell him he’s keeping a secret from no one. Synthesizer giant Roland has developed an effective pitch correction software plug-in called V-Vocal that ships with its Cakewalk Sonar audio suite. It is easy to use and operates seamlessly within the parent software. Another big player is Waves, whose Tune correction software is in use in many major studios.
                After all of the hit songs with pitch-corrected vocals came the inevitable parodies and comedy applications, most notably the "Autotune the News" team who scored a major hit Autotuning Antoine Dodson’s notorious rant on Huntsville, Alabama television following an assault attempt on his sister. It’s amazing what passes for comedy, but that’s for another essay. Somehow though, Katie Couric reading national news in a pop/R&B vocal melody on top of what sounds like a Mariah Carey rhythm bed is pretty irresistible.
                Pitch correction as a blatant vocal effect is still popular in commercial music production but seems on the wane. Pitch correction as an editing tool, however, has cemented itself forever as an essential arrow in the music producer’s quiver. It opens worlds for artists, which is precisely what music technology tools are supposed to do.