The Personalization team develops features that help listeners discover music and podcasts, including Blend and Discover Weekly. The Speak Team within Personalization focuses on creating state-of-the-art speech models for recognition, synthesis, and speech-to-speech applications with human-level emotional expressiveness.

What You'll Do

Develop and experiment with new methods for speech synthesis and speech recognition, leveraging latest research
Work on expanding speech use-cases for different markets and products
Contribute to a research team building models at scale for the Spotify platform
Champion best practices for research and development
Collaborate with engineering and data teams on infrastructure needs and pipeline improvements

Who You Are

Strong ML background with PhD plus professional experience
Experience with transformers, GANs, diffusion models, flow matching, VAEs, or audio codecs
Proven experience developing generative models for speech, audio, music, NLP, or computer vision
Strong Python and PyTorch proficiency
Excellent technical and non-technical communication abilities
Demonstrated high-quality research experience in academic or professional settings

Where You'll Be

New York-based role with flexibility for remote work and occasional in-person meetings

Senior Applied Research Scientist - Personalization