DamRsn/NeuralNote

Audio Plugin for Audio to MIDI transcription using deep learning.

What it solves

NeuralNote is an audio plugin designed to convert audio recordings into MIDI data (audio-to-MIDI transcription). It allows musicians and producers to extract musical notes from audio files or live recordings, turning them into editable MIDI notes in a Digital Audio Workstation (DAW).

How it works

The tool operates as a VST3/AU plugin or standalone app. Users can record live audio or drop in supported audio files (.wav, .aiff, .flac, .mp3, .ogg). It uses the MuScriptor transformer model (ranging from 103M to 1.4B parameters) to analyze the audio and decode it into MIDI events. The process runs entirely locally on the user's machine using the muscriptor.cpp engine, utilizing GPU acceleration (Metal on macOS and Vulkan on Windows) for faster transcription.

Who it’s for

It is primarily for music producers, composers, and musicians who use DAWs and use audio-to-MIDI transcription to capture musical ideas or potentially transcribe full mixes with multiple instruments.

Highlights

  • Multi-instrument transcription: Able to transcribe full mixes with notes grouped by instrument.
  • Fully local processing: Audio data never leaves the user's machine.
  • Model flexibility: Offers three model sizes (small, medium, large) to balance speed and quality.
  • DAW integration: Works as a plugin (VST3/AU) or standalone app, allowing MIDI to be dragged and dropped directly into a MIDI track.

Related

  • Project
    Starmel/OpenSuperWhisper

    A macOS application for real-time audio transcription using Whisper and Parakeet engines, featuring flexible keyboard and mouse triggers for quick voice-to-text conversion.

  • Project
    0xShug0/audio.cpp

    A high-performance C++ audio inference framework built on ggml that provides a portable, native runtime for running a wide variety of local audio AI models.

  • Project
    coddingtonbear/obsidian-local-rest-api

    An Obsidian plugin that runs a local HTTPS REST API and a Model‑Context‑Protocol server, letting scripts, browser extensions, and AI agents perform full CRUD, targeted markdown patches, searches, command execution, and more on your vault. Authentication is via a bearer token and a self‑signed certificate; the plugin also provides ready‑made MCP configs for Claude and Cursor.

  • Project
    pytorch/audio

    A PyTorch-based audio library that provides tools for processing audio data for machine learning, featuring GPU acceleration and specialized audio transforms.