Neural DSP’s new TINA robot uses machine-learning to model amps with “unparalleled precision”
How far have we as a society come as far as amp modelling is concerned? Not far enough apparently, if Neural DSP‘s latest innovation is any indication.
The company today announces TINA, a groundbreaking proprietary data-collection robot that promises to revolutionise the precision and authenticity of guitar amplifier modelling.
READ MORE: Dave Navarro says it was “painful” to play guitar after Taylor Hawkins died
The spaces between an amplifier’s controls — such as gain, bass, mid, treble, presence, and master — contain a vast array of complex interactions and sonic distinctions. Guitarists adjust these controls instinctively, but digitally emulating this process is challenging. Snapshot models can’t capture all the fine-tuned combinations, making it impossible to manage the millions of potential settings accurately.
Here’s where TINA, or a Telemetric Inductive Nodal Actuator, comes in. It robotically accesses the entire spectrum of every control’s range by physically connecting with those controls via actuator arms. Every control is systematically turned with its output recorded. With thousands of recorded examples, a neural network is trained to replicate the device’s behaviour for each setting.
In short, TINA combines mechanical robotics with machine learning to digitally replicate analogue devices like guitar amps – with supposedly insanely precise results.
“We’ve successfully removed all human intervention within the amplifier modelling process – ensuring an unparalleled level of precision in every model by capturing every subtle detail in the amplifier’s controls,” says Douglas Castro, CEO at Neural DSP Technologies.
“Our goal is always to create models that are indistinguishable from the real thing, even by experts,” Machine Learning Team Lead and Researcher Aleksi Peussa adds.
Image: Neural DSP
Leveraging TINA’s advanced capabilities, the firm also launched CorOS 3.0.0 and Quad Cortex Plugin Compatibility (PCOM), giving creators access to virtual devices from their Neural DSP plugins on their Quad Cortex. The initial release includes compatibility for two plugins: Archetype: Plini X and Archetype: Gojira X, with additional QC-compatible plugins set to launch in subsequent CorOS updates.
Meanwhile, CorOS 3.0.0 introduces several enhancements, including:
New devices: Transpose, Doubler, Transparent Blend, Monophonic Pitch Correction, Circular Delay, Plugin Gate, Plugin Doubler, Plugin Blend, and Plugin Graphic-9
Sidechaining added to several devices
Custom folders for better organisation of Neural Captures and Impulse Responses
Redesigned Preset MIDI Out UI
To celebrate the launch, Neural DSP is offering every Quad Cortex owner a free plugin of their choice from 31 July to 28 August.
Learn more at Neural DSP.
The post Neural DSP’s new TINA robot uses machine-learning to model amps with “unparalleled precision” appeared first on Guitar.com | All Things Guitar.
Source: www.guitar-bass.net
