Resonate

What it does

Resonate is a spectral analysis algorithm for real-time audio processing. It extracts frequency information from audio signals using a bank of resonators tuned to perceptually relevant frequencies. Each resonator accumulates signal contributions around its target frequency using an exponentially weighted moving average, producing instantaneous estimates of spectral content without buffering input samples.

Who it is for

Audio developers and researchers building real-time processing applications. The tool targets software that needs low-latency frequency analysis, such as audio effects, synthesizers, music information retrieval systems, or signal monitoring tools. It suits embedded and resource-constrained environments due to its minimal memory and computational requirements.

Pricing

The site does not show prices.

How it stands out

Resonate claims three performance advantages over conventional spectral analysis methods. First, it operates sample-by-sample without buffering, eliminating latency introduced by fixed-size analysis windows. Second, it requires minimal computation—only a handful of arithmetic operations per sample—making it suitable for real-time and embedded use. Third, it uses linear memory scaling with the number of resonators rather than input duration, keeping resource demands predictable regardless of signal length.

The algorithm builds on established phase vocoder techniques but implements them as independent parallel resonators rather than as a filterbank. This independence means resonators can be tuned to any frequencies and time constants without constraints, offering flexibility in frequency resolution and perceptual weighting.

The creator presented this work at the 2024 International Computer Music Conference, where it received the best paper award, and again at 2025 ICMC. A talk was delivered at the Audio Developer Conference in Bristol.

What a founder should check

Anyone building a competing tool should investigate three things. First, verify what existing tools dominates the real-time spectral analysis market—whether STFT implementations, commercial audio SDKs, or research projects already serve the target users and with what tradeoffs. Second, understand switching costs: audio developers often embed analysis methods deep in their pipelines, so evaluate how much friction exists around adoption and integration. Third, examine the moat: Resonate's core strength is computational efficiency and low latency. A founder should test whether these advantages remain significant as hardware improves and determine if the algorithm's perceptual tuning (rather than raw frequency resolution) is what users actually value in practice.