- Author
- Alexandre R.J. François
- Published
- Undated
- Kind
- Researcher's project page for an algorithm, with papers and audio demos
- Original
- alexandrefrancois.org/Resonate/
- Subject
- Audio models and analysis
- Themes
- What the senses do
What it is
Alexandre R.J. François, the algorithm's developer, wrote the page. It describes Resonate and an extension where resonators track frequency. The page also covers synthesis, publications and software.
What it says
- The page says each resonator updates per sample, with no buffering.
- The page says cost grows linearly with the number of resonators.
- The page says resonators can tune themselves to follow the frequencies in the input.
- The page says Resonate gives finer time resolution than FFT-based spectrograms.
- The page says the dense output can synthesize audio, with pitch and speed changes.
- The page says a live demo used 112 resonators, drawing a slice every 1.46 milliseconds.
- The page lists a Best Paper Award for the 2025 Resonate paper.
What evidence it gives
The page gives update formulas and side-by-side spectrograms against the constant-Q transform. It embeds audio of songs shifted in pitch and speed. It links to two papers and open source code. It claims real-time performance but gives no timing figures.
Interests and limits
The page is the developer's own, and it links a Resonate Store. Comparisons are shown as images, with no error measures. Performance claims are not backed by figures here. The linked papers were not read for this page.
Connections
These links are this wiki's own reading, not claims the sources make.
- How the cochlea computes (2024). The thread argues over whether the ear really performs a Fourier transform. Resonate updates a bank of resonators on each sample, with no window. The page makes no claim about ears. Still, the thread's account suggests testing it as a model of listening.