Senior Generative Speech AI Scientist

Spotify; Greater London; Hybrid; GBP 90,000 - 150,000; Full time

Job summary

Spotify is seeking a senior applied research scientist to develop novel ML methods and architectures for state-of-the-art speech-to-speech models. You will work across a full production pipeline, collaborating with engineering to improve quality and realism, and push the frontiers of speech technology.

You will join the Speak Team in London or Stockholm, with flexibility to work from home and occasional in-person meetings, contributing to research that powers end-to-end voice models for users

Qualifications

• PhD degree on top of professional experience.
• Strong ML background with experience in transformers, GANs, diffusion models, flow matching, VAEs, audio codecs.
• Experience developing generative models for speech synthesis, speech recognition, audio/music, NLP or CV.
• Strong Python skills, with PyTorch experience.
• Excellent communication and ability to present complex ideas clearly.
• Proven track record of high-quality research in academic or professional settings.

Responsibilities

• Develop and experiment with new methods for speech synthesis and recognition, including end-to-end approaches.
• Expand speech use-cases across markets and products.
• Contribute to building models at scale to power the Spotify platform.
• Champion best practices for research and development and share knowledge with Speak team.
• Collaborate with engineering and data teams to improve infrastructure and data quality for pipelines.

Skills

ML research

Transformers

GANs

Diffusion models

Flow matching

VAEs

Audio codecs

Python

PyTorch

Communication

Academic/Professional research

Education

PhD in a relevant field

Tools

PyTorch

Job description

Spotify is seeking a senior applied research scientist to develop novel ML methods and architectures for state-of-the-art speech-to-speech models. You will work across a full production pipeline, collaborating with engineering to improve quality and realism, and push the frontiers of speech technology.

You will join the Speak Team in London or Stockholm, with flexibility to work from home and occasional in-person meetings, contributing to research that powers end-to-end voice models for users