Font Sounds [Sequencer]

YouTube Uploaded by ScrollingMusic on Apr 1, 2012

"Happy April Fool's Day, ScrollingMusic style! In this silly video, we run a collage image with several different fonts through our Matlab synthesizer. (Bonus points if you can name the excerpt at the end of the video!)"

Happy Birthday Bob Moog! by ScrollingMusic

Happy (Belated) Birthday Bob Moog!

YouTube Published on May 23, 2012 by ScrollingMusic

"Bob Moog (May 23, 1934 -- August 21, 2005) revolutionized electronic music with his introduction of the Moog synthesizer line in the 1960s. This video is our tribute to his legacy, created with our own MATLAB-based software synthesizer. Happy Birthday!"

Bach: Goldberg Variations - Variation 18. Canone alla Sesta (synthesized)

YouTube Uploaded by ScrollingMusic on Feb 9, 2012

Update: Re-Uploaded by ScrollingMusic on Feb 10, 2012
[reuploaded with better balance]

"Johann Sebastian Bach (1685-1750): Goldberg Variations, BWV 988 - Variation 18. Canone alla Sexta. a 1 Clav.

this 3-voice variation is a canon at the sixth. The canon is markedly visible in the red and green voices, as the red imitates the green, delayed by a half note.

About the 18th variation:
This is a canon at the sixth in 2/2 time. The canonic interplay in the upper voices features many suspensions. Commenting on the structure of the canons of the Goldberg Variations, Glenn Gould cited this variation as the extreme example of "deliberate duality of motivic emphasis [...] the canonic voices are called upon to sustain the passacaille role which is capriciously abandoned by the bass." Gould spoke very fondly of this canon in a radio conversation with Tim Page: "The canon at the sixth -- I adore it, it's a gem. Well, I adore all the canons, really, but it's one of my favorite variations, certainly." (from Wikipedia: Goldberg Variations)

The sound featured in this video is a slightly modified version of one of our earliest MATLAB synthesizer sounds. We used our 'guitar' voice, but added slightly more harmonic content to the attack. For this video, we chose to work with a single instrument and a simple timbre to avoid an overly thick texture. While in a sense, this video is a sort of returning to our roots, it is also our first video to be rendered in 1080p :)"

C.P.E. Bach: Solfeggietto (synthesized)

YouTube Uploaded by ScrollingMusic on May 12, 2011

"From WP:

Solfeggietto (H 220, Wq. 117: 2) is a short solo keyboard piece in C minor composed by C. P. E. Bach. Played at a tempo of Allegretto (Or Presto to Pretissimo, by some), it requires well-trained, dexterous fingers and much practice to be played well and at a consistent pace. The piece is not meant to be played at a fast, rushed pace, but more an even, quick pace. It contains changing C minor arpeggios and fast 16th notes. Some pianists use Solfeggietto to warm up their fingers for a performance.

This synthesized rendition was generated from a MIDI file by our Matlab synthesizer. For this piece, we used our "amvoice" instrument. Instead of generating a unique timbre through additive synthesis, like most of our instruments do, it generates sound through amplitude-modulation (AM) synthesis, multiplying fundamental sine waves instead of adding them. This allows a harmonic-rich tone to be created in a very efficient manner, with a run-time much faster than additive synthesis instruments of a comparable tone quality."

Schubert: Impromptu, Op. 90, No. 3 (synthesized)

YouTube Uploaded by ScrollingMusic on Sep 5, 2011

"The third impromptu in the Op. 90 set is one of Schubert's popular and well-known works for solo piano. It is written in G-flat major and in 4/2 time, but when it was first published, the meter was changed to 4/4 and it was transposed up to G. This was done as the publishers thought the accidentals and unconventional meter would discourage students from learning the piece. However, this transposition actually made the piece more difficult to play, and spoiled the flat-key relationship between the impromptus in Op. 90. Today, most current published editions have restored the piece to its original tonality and time. The piece, featuring long melodic lines, showcases Schubert's excellent lyrical facility. Musical tension builds from the calm beginning of the piece to a climax in the middle, then relaxes again as the piece draws to a close.

This synthesized rendition and the scrolling visualization were created by our Matlab synthesizer from a MIDI file. The timbre used for this piece is a combination of two "instruments" from our synthesizer: a subtractive synthesis-based voice for the main melodic lines and an FM/additive synthesis voice (whose sound was actually created by accident) adding shimmery, swirling overtones to the mix."

Early experiments: encoding images with sound

YouTube Uploaded by ScrollingMusic on Mar 23, 2011

"The predecessor to our MATLAB synthesizer is a function we call RevSpectrogram, which interprets an image of any size as a spectrogram of a sound, and creates the sound based on the intensity of the pixels in the image. When the sound is visualized, the spectrogram image obtained retains much of the properties of the original image.

We encoded four sample pictures with RevSpectrogram:
1. An improvised picture with white lines
2. Picture of a gull (taken by Andy Wang)
3. "Earthrise", taken by William Anders in 1968 during the Apollo 8 mission to the moon
4. A Windows glass desktop background

The reverse spectrogram interpolates each pixel to a linearly-spaced frequency index. The frequent clicks you hear prevalently are a result of the linearly-spaced sine waves matching up every 1/20th of a second."

In Motion - The Social Network - Trent Reznor and Atticus Ross

YouTube Uploaded by ScrollingMusic on Jul 13, 2011

"Excerpt from "In Motion," featured in the 2010 film "The Social Network." Music by Trent Reznor and Atticus Ross.

The soundtrack to "The Social Network" received widespread acclaim when the film was released. It has one both the 2010 Golden Globe and Academy Awards for best original soundtrack. This particular track is featured early in the film, during the scene where Mark Zuckerberg is hacking onto the Harvard house servers and downloading pictures of undergraduate girls to create his comparison site, facemash.com.

The electronic sound of this track is well-suited to our Matlab synthesizer. This rendition closely mimics the sound of the original, but we did not design any new instruments. In fact, the mix here contains some of our earliest Matlab instruments (sax, guitar, triangle wave.) We did not use a MIDI file to sequence this track. Instead, we fed the image displayed in the video to our synthesizer, and it interpreted the "piano-roll" notation to create this rendition."

Joplin: The Easy Winners (synthesized)

YouTube Uploaded by ScrollingMusic on Oct 24, 2011

"Scott Joplin (c. 1867-1917): The Easy Winners (1901)

This ragtime composition by Scott Joplin is one of his most popular. It was first published in 1901, and is one of four Joplin pieces that had been recorded as of 1940. This synthesized rendition along with the scrolling "piano-roll" visualization were generated by our Matlab synthesizer from a midi file. The timbre used for this piece was generated through phase modulation synthesis - that is by introducing time-variation to the phase parameter ϕ in the general sinusoid equation y = A*sin(ωt + ϕ). For example, ϕ can be given by another sinusoidal function, resulting in y = A*sin(ωt + sin(ωt)). This is a computationally efficient synthesis method, allowing harmonically rich waveforms to be generated with relatively few arithmetic and trigonometric operations."

Bach: Sonata in A major, BWV 1015 - III. Andante un poco

YouTube Published on Apr 30, 2012 by ScrollingMusic

"Johann Sebastian Bach (1685-1750): Sonata No. 2 for violin and harpsichord in A major, BWV 1015 - III. Andante un poco.

This is the beautiful slow third movement of Bach's Sonata BWV 1015 for violin and harpsichord. The sonata is notable for its intricate canonic counterpoint. This movement, in particular, features a canon at the unison between the violin and harpsichord, offset by a bar, accompanied by a near-perpetually moving bass line. The canon has been visually represented by the scrolling red and green bars (red is the violin part; green is the echo at the unison by the harpsichord). The movement modulates to its dominant minor key before returning to its home tonality of F-sharp minor.

For this video, we synthesized the harpsichord part with our existing guitar and vibraphone sounds; we feature a new sound sampled from a recorder/ocarina for the violin part. The accompanying visualization consists of both a spectrogram and our usual scrolling bar visualization."

Barber: Adagio for Strings (synthesized)

YouTube Uploaded by ScrollingMusic on Feb 13, 2012

"Samuel Barber (1910-1981): Adagio for Strings

This is Barber's 1936 arrangement of the second movement of his own String Quartet, Op. 11. Barber transcribed this work once again in 1967 as an eight-part choral setting of the Agnus Dei. This piece is one of the most recognizable in classical music, having been featured in many TV shows and films. In 2004, listeners of the BBC's Today program voted this piece the "saddest classical" work ever.

This piece is made up of 10 basic phrases, with the climax residing in the fifth phrase, around 5:30 into the piece. Each phrase begins with a single pitch held by a string section (except the phrase immediately after the climax), followed by a resounding chord by the full string orchestra. In the aftermath of the long fifth statement and climax, the orchestra breaks off into short, almost incomplete thoughts. The overall tonality of the piece suggests is B-flat minor; the piece ends on a beautiful lingering unresolved V chord (F major). The melodic line is surprisingly simple and diatonically stepwise.

The awesome new sound featured in this video, developed by Drew Weymouth, is derived from filtering out a fundamental and several harmonics for each note from pure white noise (as opposed to additive synthesis of periodic waveforms). This process is computationally intensive, requiring very steep, narrow bandpass filters with passband widths of a few hertz or less to achieve a pitched sound. An FIR (finite impulse response) implementation requires several thousand filter coefficients for an appropriate rejection of undesired frequencies. Because the waveform generation is intrinsically random, the resulting output has a full, lifelike, and ethereal sound that dynamically changes even as a single pitch is sustained. The result is a powerful commanding sound reminiscent of church organ pipes mixed with low brass, strings, and light vocals."

Beethoven: "Moonlight" Sonata mov. 3 (synthesized)

YouTube Uploaded by ScrollingMusic on Jun 25, 2011

"Ludwig van Beethoven (1770-1827): Piano Sonata No. 14 in C# minor ("quasi una fantasia"), mov. 3

Composed in 1801, the "Moonlight" sonata is one of Beethoven's most famous works for piano. Beethoven included the "quasi una fantasia" ("almost a fantasy") phrase in the sonata's title because it does not follow the traditional fast-slow-fast structure of 3-movement sonatas. While the style of this third movement is typical of Beethoven's sonatas, the first movement is quiet and pensive. Fast arpeggios and chromatic runs make this movement one of Beethoven's most technically demanding piano compositions. It is also thought that this movement inspired Chopin's Fantaisie-Impromptu.

This synthesized rendition was created by mixing together two instruments from our Matlab synthesizer: guitar and a subtractive synthesis-based keyboard voice. Like an acoustic piano, the timbre of this mix changes with dynamic levels, along with the volume. The dynamics are also reflected in the "piano-roll" visualization, with blue notes being piano or pianissimo, green representing mezzo-piano and mezzo-forte, and red corresponding to forte/fortissimo. Both the audio and scrolling visualization were generated by our Matlab synthesizer from a MIDI file."

Bach: Sinfonia No. 15, BWV 801 (synthesized)

YouTube Uploaded by ScrollingMusic on Dec 8, 2011

"Johann Sebastian Bach (1685-1750): Sinfonia No. 15 from Inventions and Sinfonias, BWV 801

This piece comes from Bach's collection of Inventions and Sinfonias (also known as Two- and Three-Part Inventions). Bach composed the set as an exercise for his keyboard students, stating that students "are shown a clear way not only to learn to play cleanly in two parts, but also, after further progress, to handle three obligate parts correctly and well; and along with this not only to obtain good inventions (ideas) but to develop the same well; above all, however, to achieve a cantabile style in playing and at the same time acquire a strong foretaste of composition." This sinfonia is the last of the set, and is one of the more challenging. It requires great dexterity in the fingers to play its fast, flowing runs cleanly and articulately.

This synthesized rendition, along with the scrolling visualization, were generated by our MATLAB synthesizer. The timbre is a combination of three previously created sounds, as well as a new voice created specifically for this piece."

Rachmaninoff: Prelude in G minor, Op. 23, No. 5 (synthesized)

YouTube Published on May 18, 2012 by ScrollingMusic

"Sergei Rachmaninoff (1873-1943): Prelude in G minor, Op. 23, No. 5 (1901).

This famous prelude is the fifth of a set of ten preludes written between 1901-1903. Rachmaninoff himself premiered the piece in Moscow on February 10, 1903. The prelude is in ABA form, opening with an Alla marcia, before entering a more lyrical melancholy section, and finally concluding with a recapituation of the march.

The visualization in this video is our usual scrolling-bar visualization, with the colors of the bars corresponding to the dynamic level of the notes. The timbre of the synthesized rendition was created by mixing together several raw outputs from our MATLAB synthesizer using different "instrument" functions. The synthesis techniques used include additive, subtractive, and phase distortion."

Schubert: Ständchen

YouTube Uploaded by ScrollingMusic on Nov 1, 2011

"Franz Schubert (1797-1828): Ständchen, from Schwanengesang, D 957, a posthumous collection of songs. Ständchen ("Serenade"; the singer exhorts his lover to make him happy)

The song featured may be the most popular of Schubert's songs in this collection, which uses poems by three poets, Ludwig Rellstab (poet of this song), Heinrich Heine and Johann Gabriel Seidl.

The notes in the bass is a combination of several existing synthesized instrument sounds (clarinet, electric piano, vibraphone and guitar) with modified attack, decay, sustain, and release parameters. The sound for the voice is also a mix of sustaining instruments: clarinet, saxophone, organ, and vibrato wind to name a few.

We used a scrolling spectrogram, as the graphic plot of the spectral content of an audio signal. The vertical axis represents frequency, and the horizontal axis time. Higher-pitched notes appear higher up on the screen, and the harmonics for each individual note are visible above the fundamental at regularly spaced intervals."

Chopin: Mazurka, Op. 6 No. 1 (synthesized)

YouTube Uploaded by ScrollingMusic on Aug 25, 2011

"Frédéric Chopin (1810-1849): Mazurka, Op. 6 No. 1

Chopin composed at least 69 mazurkas between 1825 and his death in 1849. These short pieces for solo piano are based on the traditional Polish dance known as the mazurka. However, Chopin transformed his mazurkas into an entirely new genre, which became known as a "Chopin genre." This piece is in the key of F-sharp minor, and was the first mazurka to be published, although two were composed earlier (published posthumously.)

This synthesized rendition and the scrolling visualization were generated from a MIDI file by our Matlab synthesizer. This rendition is significant, because it achieves its timbre by combining three synthesis types: additive, subtractive, and sample-based. While the additive and subtractive voices are completely synthesized from scratch, we also added recorded samples of an upright piano (sample-based synthesis) into the mix. The sample bank consist of a recording of each of the 88 keys played in isolation."

Bach: Chorale Prelude 'Wachet Auf' (synthesized)

YouTube Uploaded by ScrollingMusic on Oct 16, 2011

"Johann Sebastian Bach (1685-1750): Wachet auf, ruft uns die Stimme, BWV 140, 645

This famous piece by Bach is the fourth movement of his church cantata "Wachet auf, ruft uns die Stimme" (also known as Sleepers Wake), BWV 140. Bach later transcribed this movement for organ (BWV 645). The chorale was first performed in 1731, and is based on the Lutheran hymn of the same title.

This synthesized rendition and scrolling visualization were generated by our Matlab synthesizer. The timbre of this piece is based on slightly modified versions of existing additive synthesis instruments for our Matlab synthesizer, and showcases an entirely new sound composed of filtered sawtooth waves."

Faurè: Après un rêve (synthesized)

YouTube Uploaded by ScrollingMusic on Jun 12, 2011

"Gabriel Fauré (1845-1924): "Après un rêve" (No. 1) from Trois mélodies, Op. 7

One of Fauré's most popular pieces, "Après un rêve" is the first piece in his suite for voice and piano, Trois mélodies, which was composed between 1870 and 1878. According to Wikipedia, "In Après un rêve, a dream of romantic elopement with a lover, away from darkness, and towards an awakening light is described. However, the dreamer longs to return to the "mysterious night". The text of the poem is an anonymous Italian poem translated to French by Romain Bussine."

This synthesized rendition was created by our Matlab synthesier, which interpreted a MIDI file to create both the audio and the scrolling visualization. The mix for this piece was created by synthesizing the MIDI multiple times, played by different instruments we have programmed, and mixing the results together. It contains a flute-like vibrato-based timbre, along with guitar (one of our earliest instruments,) and a newly created "fuzz/distortion" sound."

Matlab Synthesizer - Beethoven: "Waldstein" Sonata

YouTube Uploaded by ScrollingMusic on Mar 3, 2011

"Ludwig van Beethoven (1770-1827): Piano Sonata No. 21 in C major, Op. 53. I. Allegro con brio

This sonata is considered to be one of Beethoven's greatest and most recognized, and one of the three particularly notable ones of his middle period. Beethoven completed this sonata, one of his most technically challenging works, in the summer of 1804. During this time, Beethoven had begun to develop hearing problems that would lead to his deafness in later life. The "Waldstein" name comes from the sonata's dedication, to Count Ferdinand Ernst Gabriel von Waldstein of Vienna, a close personal friend of Beethoven's.

This synthesized rendition was generated by our Matlab synthesizer. For this piece, we programmed a very velocity-sensitive sound whose harmonics change with changing dynamic levels. Volume levels alone are not enough to convey the vast scope and grandiosity of this work. The colors in the scrolling visualization correspond to dynamic levels as well, with red being the loudest notes in the piece."

CLICK FOR MORE »

Chopin: Étude Op. 10, No. 1 (synthesized)

YouTube Published on Sep 3, 2012 by ScrollingMusic

Matlab software synthesizer.

"Frédéric Chopin (1810-1849): Étude in C major, Op. 10, No. 1 (1829).

This étude is the first of a set of 12 (called Opus 10), published in 1833. It is particularly famous for its study in right-hand arpeggios often ascending and descending more than four octaves per chord. Running sixteenth notes dominate the texture of the right hand, supported by heavy octaves in the left. The study is in ternary form (ABA), and features some remarkable instances of transposition in its middle section. A harmonic reduction of this study usually leads to a smoothly step-wise 5-voice texture, reminiscent of Bach's C major Prelude from his first book of the Well-Tempered Clavier. Vladimir Horowitz considered this Chopin étude his most difficult.

The principal timbre in this synthesized rendition is generated through phase distortion synthesis - that is modulating the phase component of a standard sinusoidal wave with time. This technique creates a rich harmonic texture quite efficiently compared to additive synthesis. We programmed this phase distortion "voice" to generate more harmonic distortion as the fundamental frequency decreases. Therefore, the low octaves in the left hand have far more harmonics than those at the top of the right hand arpeggios.

The visualization is a combination between our standard scrolling "piano-roll" score and a third-party spectrogram visualization (a graphical plot of harmonic content vs. time in the.synthesized audio)."

Pachelbel: Canon in D (synthesized in just intonation)

YouTube Published on Apr 24, 2012 by ScrollingMusic

"Johann Pachelbel (1653-1706): Canon in D

This very famous work from the Baroque period was actually forgotton for centuries, rediscovered only recently in the early 1900s. Today, it has become quite ubiquitous, and is frequently performed at weddings and included in classical compilation discs. This piece combines the musical devices of canon and ground bass. Three voices are engaged in canon, where each part plays the same phrase offset by a certain number of beats. The fourth voice is a basso continuo, playing the same two-measure phrase throughout the piece. The chord progression in this piece: I - V - vi - iii - IV - I - IV - V is present in other classical works as well as much of the popular music we hear today (U2's "With or Without You" and Blues Traveler's "Hook", to name a few).

This rendition is synthesized using the "strings", "harpsichord" and "voice" instruments of our MATLAB-based synthesizer. It is tuned in just intonation, centered on D = 294 Hz. Just intonation is a tuning system in which the frequencies of the twelve chromatic pitches are chosen based on fixed ratios compared to the center pitch. For example, the frequency of the major 3rd is 5/4 times that of the root. Thus, in this case F# is 367.5 Hz. In just intonation, intervals tend to sound more consonant then in the most common equal temperament system, but if a piece modulates to a new key area, the consonant ratios between frequencies are no longer preserved."