Native Instruments

Native Instruments deployed an AI-powered Preset Explorer in Absynth 6 (released December 2025) that uses a deep learning model to analyze the timbral characteristics of every synthesizer preset and arrange the entire library into a visual map, allowing musicians to browse sounds by feel rather than scrolling through text-based lists.

The Preset Explorer was developed by Native Instruments' Applied AI team, led by Head of Applied AI Andy Sarroff. A deep-learning model analyzes each preset's 'timbral DNA' — including perceptual qualities such as brightness, sustain, and noisiness — and positions presets spatially on a visual map based on sonic similarity. This removes the need for text or tag-based searching. Simple filters allow users to narrow their path through the collection. According to the company, the system was designed to support musical exploration rather than automate creative decisions.

Native Instruments offers iZotope Nectar 4, a vocal mixing plugin that uses AI to analyze a user's vocal recording and automatically suggest processing settings — including EQ, compression, and pitch — and to generate synthesized backing vocal parts from a single singer's voice.

Nectar 4's Vocal Assistant listens to an audio input, analyzes dynamics, pitch register, and tonal characteristics, then sets EQ, compression, and other module parameters as a starting point for vocal mixing. The Backer module uses machine learning to generate alternative vocal personas from a single lead vocal, producing virtual background singers with different tonal qualities. Users can also import a reference vocal file (via the Audiolens tool) and the AI will match the tone and EQ of that target voice.

Native Instruments tested a prototype AI-powered chat interface for searching and discovering sounds in its digital library, built using Amazon Bedrock and large language models from Anthropic, designed to understand natural-language and social-context search queries from music creators.

Native Instruments' research engineering team, led by Director of Research Engineering Andy Sarroff, partnered with consultancy Tribe AI in a four-week proof-of-concept engagement. The prototype combined Native Instruments' existing indexed sound file data (tagged audio files) with a generative AI layer powered by Anthropic's large language models via Amazon Bedrock. The goal was to enable the search interface to understand nuanced, colloquial, or pop-culture-reference-based queries (e.g., a user asking for sounds that evoke a specific feeling or era). As of the case study's publication, the prototype had been shared internally and was entering an A/B testing phase with company leadership; public deployment had not been confirmed.