PhD Research Internships – Music Generation / Source Separation and Enhancement / Music Information Retrieval
About Us
Moises is a fast‑growing startup in the Music Tech space, building next‑generation tools that empower musicians and producers. We combine cutting‑edge machine learning, audio signal processing, and deep music domain knowledge to develop tools that improve music creation, mixing, performance, and analysis.
Overview
We are looking for motivated research interns to join our team. The roles focus on either:
- Audio Source Separation and Enhancement
- Music Information Retrieval (MIR)
- Music Generation
You can apply to one or multiple roles, depending on your background and interests.This internship is strongly oriented toward PhD candidates currently enrolled in relevant programs, though exceptional MSc candidates with strong research experience may also be considered. Role 1: Source Separation and Audio Enhancement Intern - Responsibilities:
- Research and prototype models for isolating instruments, vocals, and other musical components.
- Explore audio enhancement approaches such as denoising, dereverberation, and quality restoration.
- Work with large audio datasets, model training pipelines, and evaluation metrics.
- Collaborate with engineers and audio specialists to integrate models into music‑production‑oriented workflows.
- Preferred Qualifications:
- Background in audio signal processing, machine learning, or related fields.
- Experience with deep learning frameworks such as PyTorch.
- Familiarity with source separation techniques.
Role 2: Music Information Retrieval (MIR) Intern - Responsibilities:
- Develop algorithms for tasks such as beat tracking, chord recognition, structural segmentation or tagging.
- Experiment with machine learning and signal processing approaches to extract insights that support musicians and producers.
- Work with the team to integrate MIR features into creative and production‑oriented tools.
- Preferred Qualifications:
- Background in MIR, audio analysis, and machine learning.
- Experience with deep learning frameworks such as PyTorch.
- Musical intuition, theory knowledge, or personal music production experience is a strong plus.
Role 3: Music Generation Intern - Responsibilities:
- Research and prototype generative models for conditional music generation.
- Experiment with diffusion and/or autoregressive models, and embedding and/or token‑based audio/music representations.
- Collaborate with the team to integrate generation features into tools intended for musicians and producers.
- Preferred Qualifications:
- Background in deep learning, generative modeling, computational creativity, or music technology.
- Experience with deep learning frameworks such as PyTorch.
- Strong musical intuition or experience in composition or production is a significant plus.
General Internship Details
- Flexible start date
- Fully remote
- Full‑time or part‑time options available