LP

    LSTM Piano MIDI Generator

    1

    Generate original piano MIDI sequences using a pre-trained LSTM neural network. Customize note count, sampling temperature, tempo, and output location with a simple Python-based workflow.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more
    LP

    LSTM Piano MIDI Generator

    Example session with this skill installed

    Generate a 150-note piano MIDI file named 'ambient_idea.mid' using a temperature of 0.7. Download the model files if they aren't here.

    • Read your context and instructions
    • Compiled the lstm piano midi

    I have initialized the music generation sequence using the LSTM model. I verified the local environment and downloaded the necessary music_lstm.keras and pitches.npy assets from the repository.

    MIDI Generation Report

    ParameterValue
    Output Fileambient_idea.mid
    ModelLSTM (Keras)
    Note Count150
    Temperature0.7
    Tempo100 BPM
    Time Signature4/4
    Note Duration0.5 (Fixed)
    StatusSuccessfully Exported

    The generated file ambient_idea.mid contains a sequence of 150 notes. Each pitch was predicted based on the previous 50-note window context. Note that the MIDI uses a fixed velocity and rhythm; you will need to apply humanization or rhythmic quantization in your DAW for varied timing.

    Next steps

    1. Import ambient_idea.mid into your preferred MIDI synthesizer or DAW (e.g., Ableton, FL Studio, or Logic Pro).
    2. Assign a high-quality VST instrument to the MIDI track to hear the generated melody.
    3. Quantize or adjust note lengths if you require varied rhythmic patterns.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    LSTM Piano MIDI Generator

    Generate original piano music sequences using a pre-trained Long Short-Term Memory (LSTM) neural network.

    This skill helps you generate MIDI music through a Python-based workflow. It uses a trained deep learning model to predict musical notes sequentially and create new piano compositions.

    Features

    • AI Music Generation: Generate piano MIDI sequences using an LSTM neural network.
    • Customizable Generation: Adjust the number of generated notes and sampling temperature.
    • MIDI Output: Save generated compositions as standard MIDI files.
    • Python-Based: Uses TensorFlow, Keras, NumPy, and music21.
    • Local Execution: Generation runs on your own computer rather than requiring a paid cloud music-generation API.
    • Simple Workflow: Run the generation script through your compatible AI agent.

    How It Works

    1. The skill loads the pre-trained LSTM model and its pitch vocabulary.
    2. It uses a sequence of musical notes as input.
    3. The model predicts subsequent notes using probabilistic sampling.
    4. The generated sequence is converted into a piano MIDI file.
    5. The output is saved to your selected location.

    Requirements

    • Python 3.13.9
    • TensorFlow 2.21.0
    • Keras 3.15.1
    • NumPy 2.5.4
    • music21 10.5.1

    A compatible Python environment is required. The model and vocabulary are downloaded from the creator's GitHub repository when needed.

    Important Notes

    • This skill generates symbolic MIDI sequences, not rendered audio files.
    • MIDI playback requires a compatible MIDI player or digital audio workstation (DAW).
    • Results depend on the trained model and its learned musical patterns.
    • Generated compositions may require further editing for musical quality.

    Project

    Developed as part of an Artificial Intelligence internship project.

    Developer: Noorwali Khan

    Source Code: https://github.com/Noor72353/CodeAlpha-Internship

    Technology: Python | TensorFlow | Keras | LSTM | NumPy | music21 | MIDI

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.

    Generates pitches with fixed durations only. No support for chords, velocity, or timing. Requires manual setup of model assets and Python 3.13.9 environment.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 5 days ago

    • Free with an Agensi account

    Accounts
    GitHub
    Software
    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access

    Allowed Hosts

    raw.githubusercontent.com
    Listed5 days ago

    What's inside

    Frequently Asked Questions

    This skill generates a raw MIDI (.mid) file containing a sequence of piano pitches. You can drag and drop this file directly into any Digital Audio Workstation (DAW) like Ableton, FL Studio, or Logic Pro to trigger your own virtual instruments.

    Unlike general-purpose models that often struggle with file formatting, this skill uses a specialized Long Short-Term Memory (LSTM) network trained exclusively on melodic transitions. This ensures every output is a structurally sound MIDI file rather than just a text description of music.

    The skill handles the heavy lifting by automatically downloading the necessary Keras model weights and pitch vocabulary files upon your first run. You only need to ensure you have Python 3.13.9 installed along with the standard MIDI libraries mentioned in the requirements.

    Yes, you can adjust the "Temperature" parameter to control the creativity of the output. A lower temperature produces more predictable, traditional melodies, while a higher temperature encourages the LSTM to take more experimental risks with pitch selection.

    The current version focuses on monophonic piano melodies with fixed durations. It is designed to provide melodic "seeds" and does not currently support polyphonic chords, volume (velocity) dynamics, or multi-instrument orchestration.

    Free