Audio Machine Learning Engineer (Machine Learning Developer)

Role overview Develop machine learning systems for singing voice synthesis and voice conversion, from early research through production deployment. You’ll work with an AI team to improve generated vocal quality and reliability, using large music and vocal datasets and modern audio modeling methods.

Responsibilities - Research, train, and improve models for singing voice synthesis and voice conversion. - Explore diffusion-based approaches to generating vocals and audio. - Develop and work with neural vocoders, including GAN- and diffusion-based designs. - Build and refine audio feature extraction pipelines for vocal modeling. - Work with large, high-quality vocal and music datasets. - Improve model quality, robustness, and inference speed, then integrate changes into production systems. - Write efficient, maintainable Python code using PyTorch.

Requirements - Master’s degree or higher in machine learning, artificial intelligence, computer science, or a related discipline. - Strong experience developing machine learning systems, with Python as a primary language. - Hands-on experience training and deploying deep learning models with PyTorch. - Solid understanding of singing voice synthesis, voice conversion, and current audio modeling methods. - Familiarity with diffusion-based generative models and neural vocoders. - Strong written and spoken English, with the ability to take ownership and collaborate effectively in a remote team.

Nice to have A background or strong interest in music production, vocals, or audio engineering.

Benefits and work setup Full-time, fully remote work. Candidates must be based within GMT−3 to GMT+4. Compensation is described as competitive and based on experience.