Skip to content
View syrusaslam's full-sized avatar
3 followers · 4 following

Achievements

Achievement: Pull Shark

Achievements

Achievement: Pull Shark

Highlights

  • Pro

Block or report syrusaslam

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
syrusaslam/README.md

Hi, I'm Syrus!

  • I’m currently an Applied Math and Computer Science student at UC Berkeley.
  • I have a passion for AI, machine learning, and software development.
  • Below are some of my projects and contributions:

Projects

Melodify AI

  • Description: Built an Ableton Live plugin that reads project-wide MIDI data and uses a pre-trained transformer neural network to generate melodies and chord progressions, seamlessly integrating new compositions within the Ableton UI. Utilized Max for Live and Ableton's API to rapidly prototype and deploy, enabling direct MIDI manipulation through JSON and UDP communication for real-time updates, minimizing disruptions to the creative flow.
  • Tech Stack: Python, JavaScript, Hugging Face.

Predicting Volcanic Activity Using Infrasound Data

  • Description: Developed an infrasound monitor using Raspberry Pi to detect seismic and atmospheric activity through NASA California Space Grant Consortium. Implemented k-means and agglomerative clustering and neural networks using scikit-learn and TensorFlow with over 25 million values to analyze infrasound wave patterns for predicting volcanic activity with over 90% accuracy. Preprocessed data and enhanced the infrasound machine learning model, boosting accuracy by 20%.
  • Tech Stack: Python, TensorFlow, scikit-learn, pandas, numpy, obspy.

AI-based Skin Disease Detection

  • Description: Developed a convolutional neural network (CNN) using PyTorch for skin disease detection, achieving over 90% accuracy in classifying 8 types of skin diseases.Implemented a user-friendly web interface for seamless image uploads and real-time AI-driven disease predictions.
  • Tech Stack: Python, PyTorch, React.js, HTML, CSS.

Music Genre Classification and Recommendations Using Machine Learning Models

  • Description: Developed a music genre classification system using random forest decision trees to accurately categorize new songs based on their audio features with over 80% accuracy. Utilized scikit-learn, librosa, and pandas for efficient feature extraction, data processing, validation testing, and model training, ensuring high performance and reliability in music genre prediction.
  • Tech Stack: Python, librosa, pandas, scikit-learn.

Get in Touch

Pinned Loading

  1. arnavsurve/music-genre-classification arnavsurve/music-genre-classification Public

    Audio similarity experiments with SKLearn random forest classification

    Python 3

  2. arnavsurve/ARC-NASA-CSGC-infrasound arnavsurve/ARC-NASA-CSGC-infrasound Public

    Research under Professor Paulo Afonso

    Python 1

  3. melodify-ai/.github melodify-ai/.github Public archive

    1

  4. AI-Skin-Disease-Detection AI-Skin-Disease-Detection Public

    Python 1

  5. syrusaslam syrusaslam Public