Introducing PIANO FORTE, a physical piano sound source model based on AI technology
Today I would like to introduce PIANO FORTE, a physical model sound source.
Since it is not sampling, the capacity is small, but the CPU load is very high.
It is not a sound source that can be recommended at the moment.
It has been released as open source on GitHub, so please look forward to future updates.
Sound demo
With a notebook PC, the CPU goes up to 3% just by playing about 100 sounds, so I couldn't prepare a sound demo.
I may add it at a later date.
Download
You can download it without registration.
Go to the above link and click the appropriate format link from "Assets".
Usually the VST3 version (PianoForte.vst3) is OK.
Installation
Downloaded file (PianoForte.vst3) into the VST3 folder below.
Overview
We are happy to announce the release of Piano Forte, our free open source piano plugin. Available in VST3, standalone Windows (x64) and LV2 formats, this plugin uses a dual engine that blends lightweight neural networks and physics models to ensure the best balance between efficiency and fidelity. I'm here.Modeled to capture the best features of famous models such as the Steinway & Sons Models B and D and the Yamaha C5, we aimed for a well-proportioned, flexible sounding instrument.Please download it, use it, and let us know what you think.
About Omnessonos
We are a small team of music-loving developers led by Carlos Tarjano who spend their spare time researching and developing virtual instrument plugins.Below are some publications.A generic algorithm for real-time emulation of pitched instruments and vocals.Federal University of Fluminense, 2022, Doctoral Dissertation in Industrial Engineering
An Efficient Algorithm for Decomposing Quasi-Periodic Digital Signals into Pseudo-Periodics: An Application to Lossy Audio Compression.IEEE/ACM Transactions on Audio Speech and Language Processing, Vol.30, May 2022.
Envelope estimation using geometric properties of discrete real signals.Digital Signal Processing, Volume 120, January 2022.
Neurospectral Speech Synthesis: Utilizing properties of the discrete Fourier transform in real-time simulation of musical instruments using parallel neural networks. Lecture Notes in Computer Science, Volume 11730, September 2019.
Summary
Comment