PRISM

Performance Restoration & Isolated Structural Modeling

PRISM
Performance Restoration & Isolated Structural Modeling

Meet PRISM

PRISM evolves the practice of music production with a powerful new methodology for editing music and mixing sound. It reimagines the process from the user’s perspective, and reinvents it from a technical one. Yet it retains compatibility with existing tools and workflows, and preserves granular control. PRISM transitions DAWs to VAWs: Virtual Audio Workstations.

A computer rendering of the control room in a recording studio.

PRISM‘s “audio to MIDI to audio” gives users extraordinary new abilities. But it’s not a feature—it’s one of several underlying technologies used by PRISM to create the real magic. And bridge the gap between audio data and LLMs. Since 2013, PRISM has been used behind-the-scenes with dozens of iconic artists for whom performances are critical—ranging from Alice Cooper and Deep Purple, to Steve Vai and Crosby, Stills & Nash. Ten years later, PRISM is going public.
Year: 2020 Chad Wackerman & John Ferraro: Drums / Steve Morse & Blues Saraceno: Guitar / Sterling Ball: Bass

Meet the Scientist

Dr. Bill Evans seated in front of a mixing console with the native Belexes Virtual Audio Workstation interface in front of him.

Bill Evans performed his graduate work at Manchester Metropolitan University and Glasgow University. His PhD thesis established the methodology of Performance Restoration, a set of technologies and techniques to repair musical errors in recorded music, and restore performers’ original intentions. The Virtual Audio Workstation was introduced in Evans’ alternate thesis. Based on this research, he has engineered tracks for dozens of iconic artists, including with his own band, Flying Colors.

Bill Evans is a brilliant audio scientist who creates highly innovative and genuinely musical software solutions to some of recording’s most vexing issues.
Bob Ezrin
Producer/Legend
Bill Evans is one of those rare minds who bridges deep technical mastery with a true musical sensibility. His work as an audio scientist consistently pushes boundaries, not just solving problems but doing so in ways that respects the integrity of the music.
Jim Odom
Founder/CEO Emeritus, Presonus/Fender
Bill continues to make impossible things happen. PRISM converted my guitar parts to MIDI, and back to audio, preserving the performance perfectly. That just doesn't happen.
Steve Morse
7-time GRAMMY nominee, guitarist and songwriter for Deep Purple, Kansas, Dixie Dregs
Technology so easily robs us of originality, and diminishes our creativity. Bill's work has the extraordinary potential to make us more musical…and most notably, more human.
Peter Collins
Producer (Elton John, RUSH, Bon Jovi)

Foreseen Consequences

PRISM allows acoustic audio recordings to be edited as a series of independent, temporal events. Each event is a superposition of every possible performance of that note. By analysing the surrounding events, the context of any specific event can be sufficiently determined to enable the corresponding audio to be determined, and then synthesised.

A wide photograph of the Rosetta Stone. A wide photograph of the Rosetta Stone, modified to show the three representation of music within PRISM.

A subset of this process can be brought to Digital Audio Workstations. An encoding process analyses the recorded audio. It assigns semantic meaning and hierarchical representation to represent musical data, producing a MIDI track. And modelling data to represent the range of possible sounds, based on each note’s performance context. A PRISM-enabled Virtual Instrument then renders the sequence—or any version thereof—in real-time.

Year: 2020 Chad Wackerman & John Ferraro: Drums / Steve Morse & Blues Saraceno: Guitar / Sterling Ball: Bass

Foreseen Consequences

A problem manifests when raw audio is edited as MIDI. When notes are pulled apart, they expose Theoretical Audio—the parts of notes cut off by the next ones. PRISM generates the audio that would have existed, according to the performance parameters it predicts.

Split Personality
Year: 2025 “Apache” Audio Loop

PRISM divides acoustic recordings into performances and the sounds they made. Each can be edited independently using standard DAW tools (for processing MIDI and audio). Together, new classes of sound transformation are enabled to solve canonical problems with traditional music production.

A photo of a white, metal snare drum. An architectural-style diagram of a snare drum.

In this video, PRISM translates an audio loop to MIDI. Using features of PRISM’s Performer and Transformer system, the performance, instrumental physics, ambience, and other elements are edited separately. The DAW‘s native editing capabilities are complemented by PRISM‘s own unique transformations.

Third Degree Burn

PRISM‘s third iteration strategically integrates AI according to a roadmap begun in 2015, and concluding in 2029. While other systems are just discovering generative sound, PRISM has been creating hits with it for over a decade.

2015
⏹️
2018
1️⃣
2024
2️⃣
2026
3️⃣

Features that are new to the industry have been evolving strategically in PRISM, so the larger technical landscape of 2026 plugs right into PRISM where it was intended. Our training data isn’t harvested—PRISM audio can be copyrighted. And artists control the use of their likeness, receiving royalties for its use.

Meanwhile, everyone else seems to be inventing things as they go along…like that one that rhymes with, you know.

Magic. Promptly.

PRISM Version 2 added the unprecedented ability to losslessly convert audio to MIDI.  Version 3 integrates traditional, prompt-based genAI audio tasks into PRISM’s decoding process, enabling the unique synergy of absolute control to generative audio.  

1

File Under Everything

The Encoder provides the universal entry point for audio files into PRISM.
2

Make Arrangements

PRISM's Odometer composes compelling arrangements (e.g. strings) based on the music in the input audio file. Users describe their arrangement with standard text prompts.

3

A Masterpiece in Paint

Audio inpainting provides holistic, contiguous changes based on user descriptions. These changes can then be refined with absolute precision using PRISM's MIDI conversion and virtual instrument.

4

Cover Your Bass

PRISM's take on Covers enables instruments to be replaced, as well as genres. As with all of PRISM's synthesis, all of the audio data is losslessly converted to MIDI, with playback through PRISM's dynamic virtual instrument.

5

Sounds Flexible

After processing and analysing an audio file, PRISM reconstructs the performance as either rendered audio or lossless MIDI with a generative virtual instrument.

High Profile

PRISM represents all performances as discrete events, described by high-level meta data that parameterises many characteristics of human performance. PRISM’s Personality Profiler combines the power of LLMs with the specificity of MIDI.

1

Make No Mistake

Based on Evans' 2018 PhD thesis, PRISM automatically corrects users' performance errors. It is the opposite of quantizing; PRISM's Virtuoso module corrects errors musically, based on stylistic elements of artistic performance.

2

Best in Show

PRISM's Actualiser module makes performance error-correction personal, by repairing musical mistakes according the user's own performance style. It helps fulfill the ambition of all musicians—to show the best version of themselves.

3

Featured Guests

Evans pioneered stylistic imprinting in 2015 on the landmark album, by correctly predicting a missing 7-second gap in Steve Morse's guitar solo. PRISM's Influencer module generalises this practice, allowing artists to license their likeness. Users can apply a gradient to control how much of their own performance is transformed as the selected artist.

4

Inspired Compositions

PRISM's Composer module provides generative composition in the style of the user's own repertoire—or that of licensed artist.

Selling Out

Hear PRISM on tracks with iconic artists, engineers and producers, engineered by Evans. Live songs are from the original audio—no overdubbing, re-recording or sample-layering.

Steve Vai ▪ Steve Morse ▪ Mike Portnoy ▪ Neal Morse ▪ Alice Cooper ▪ Jay Graydon ▪ David Foster ▪ Van Romaine ▪ Albert Lee ▪ Sterling Ball ▪ Joe Bonomassa ▪ Angel Vivaldi ▪ Chad Wackerman ▪ John Ferraro ▪ Marco Minneman ▪ Jim Cox ▪ Steve Lukather ▪ Crosby, Stills & Nash ▪ Peter Frampton ▪ Peter Collins ▪ Ken Scott ▪ Bob Ezrin ▪ Michael Brauer ▪ Mark Neeham ▪ Howie Weinberg ▪ John Petrucci ▪ Deep Purple

PRISM Songs/Albums

27 Videos
Information
Information

Year: 2021

PRISM Usage: Steve’s guitar

Stay in the Loop

You will first receive an email to confirm you signed up. We do not use email for commercial purposes, or share with third parties.

Stay connected with PRISM, and get an inside look at the sessions from some of music’s most compelling artists. New technologies while they’re in the incubator. And fresh takes on how the science of music production affects us as engineers, producers, artists and humans.

Sound Evidence

It is not enough to create believable audio here. Longer and shorter notes are performed differently. And human audio perception is highly-tuned to the contextual interaction of simultaneous instruments. PRISM incorporates these elements into Theoretical Audio—independently of ambience.

Traditional GenAI creates believable sounds, but we need a higher standard to accurately convey performances. If the audio never existed, though, how do we know if PRISM gets it right?

Year: 2022 Marco Minnemann: Drums

    GDPR and CPRA compliance: We do not use email for promotional or commercial activity, or share them with third parties.

    CeltiphoniX

    Management + Publicity
    Go
    Bill is represented by Jeremiah Graham at Celtiphonix.

    You’ll receive an email to confirm you signed up. Your address is protected as per our privacy policy.

    Stay connected with PRISM, and get an inside look at the sessions from some of music’s most compelling artists. New technologies while they’re in the incubator. And fresh takes on how the science of music production affects us as engineers, producers, artists and humans.