Research Scientist in Generative Audio at Spotify (London or Stockholm)

Job Description

Spotify is hiring research scientists in generative audio, at all levels of seniority, for its Artist-First AI Music Lab, working on diffusion and flow-matching models for music.

At a glance

• Position: Research Scientist – Generative Audio (all seniority levels)
• Where: London or Stockholm; the posting allows work from EMEA locations where Spotify has an office
• Pay: Not stated in the posting
• Type: Permanent, full time
• Core hours: 15:00–18:00 CET
• Apply: open until filled (re-checked 13 November 2026)

About the generative audio research scientist role

• Research generative audio with diffusion or flow-matching models in one or more areas: vocal and speech synthesis; post-training for music generation; or music and audio editing
• Post-training work covers preference alignment (such as DPO, RLHF or KTO), reward-model design and reinforcement learning for quality and controllability
• Editing work covers iterative generation such as stem replacement, changes of instrumentation, mood, tempo and structure
• Run large-scale experiments on Spotify’s infrastructure and turn research into product features
• Work in a cross-functional team with engineers, designers, product managers and user researchers
• Publish, give talks and attend leading conferences

Spotify says the lab’s products follow four principles: upfront agreements with record labels, distributors and publishers; artists and rights holders choosing whether to take part; fair pay and credit for rights holders; and tools that strengthen, rather than replace, the link between artists and fans.

Why this role matters

Generative music raises hard questions about rights and consent, and this lab is built around licensed partnerships and artist opt-in. That makes it one of the few places where audio researchers can work on state-of-the-art music generation for a listener base of hundreds of millions, while working within agreements with the music industry.

What Spotify is looking for

• A PhD in computer science, mathematics, engineering or a related field; industry experience helps
• Experience in generative modelling, machine learning, music information retrieval, speech or audio processing, signal processing, probabilistic modelling or computer vision
• Deep expertise in at least one focus area: vocal or speech synthesis, post-training alignment (PPO, GRPO, DPO), or audio-to-audio generation and text-guided music editing
• Publications at venues such as ICASSP, ISMIR, INTERSPEECH, ICLR, NeurIPS, ICML, AAAI or CVPR
• Strong Python, PyTorch and NumPy skills, and the ability to explain complex topics simply

Pay and location

Spotify does not state a salary in this posting. The role is permanent; the listed locations are London and Stockholm, and the posting says you can work from any EMEA country where Spotify has a work location, collaborating in Central European and UK time zones.

How to apply

Apply through the official Spotify job posting. Spotify’s posting does not give a closing date, so the role is open until filled; ResearchJobs.in will re-check this listing on 13 November 2026. Check the posting for location and work-authorisation details before you apply.

See all our artificial intelligence jobs, or browse more research jobs on ResearchJobs.in.