Research Scientist, Generative Audio & Vocal Synthesis

Spotify; Greater London; Remote; GBP 110,000 - 170,000; Full time

Job summary

Spotify is seeking Research Scientists across levels to join its Artist-First AI Music Lab. You’ll conduct groundbreaking research in generative audio, focusing on diffusion or flow matching, vocal synthesis, post-training alignment, and audio editing to create new listening experiences.

You’ll run large-scale experiments with Spotify’s infra, publish findings, and collaborate with engineers, product managers, designers, and researchers to translate ideas into scalable products.

Qualifications

• PhD in Computer Science, Mathematics, Engineering, or related field.
• Publications in leading venues (ICASSP, NeurIPS, ICML, etc.) are a plus.
• Experience with generative modeling and ML is expected.

Responsibilities

• Conduct groundbreaking research in generative audio using diffusion or flow matching.
• Run large-scale experiments leveraging Spotify's infrastructure.
• Publish findings and present at conferences.
• Collaborate with cross-functional teams to apply research to products.
• Contribute to tools and services used by millions of users.

Skills

Problem solving

Communication

Education

Ph.D. in Computer Science/Mathematics/Engineering

Tools

Python

PyTorch

NumPy

Job description

Spotify is seeking Research Scientists across levels to join its Artist-First AI Music Lab. You’ll conduct groundbreaking research in generative audio, focusing on diffusion or flow matching, vocal synthesis, post-training alignment, and audio editing to create new listening experiences.

You’ll run large-scale experiments with Spotify’s infra, publish findings, and collaborate with engineers, product managers, designers, and researchers to translate ideas into scalable products.