Research Scientist - Generative Audio
New
S
SpotifyMusic Technology
Within the EMEA region as long as we have a work location, Central European and GMT; Core working hours are CET 3pm-6pm / EST 9am-12pm.Full-Time
Salary not disclosed
Apply NowOpens the employer's application page
Job Details
- Required Skills
- PythonMachine LearningNumpyPyTorch
Requirements
- Ph.D. in Computer Science, Mathematics, Engineering, or a related field.
- Experience in generative modeling, machine learning, music information retrieval, speech, audio, or signal processing.
- Deep expertise in vocal/speech synthesis, post-training alignment (e.g., PPO, GRPO, DPO), or audio-to-audio generation and text-guided music editing.
- Strong programming proficiency in Python, PyTorch, and NumPy.
- Proven track record of research publications at leading venues such as NeurIPS, ICML, ICASSP, ISMIR, or INTERSPEECH.
- Ability to turn research ideas into scalable, real-world product applications.
- Excellent communication skills for explaining complex technical topics.
- Demonstrated ability to build relationships with colleagues and stakeholders.
Responsibilities
- Conduct groundbreaking research in generative audio using diffusion or flow matching models.
- Research in vocal and speech synthesis, post-training alignment techniques, or iterative music generation and editing.
- Run large-scale experiments utilizing extensive infrastructure and user data.
- Create practical applications that push the boundaries of music listening experiences.
- Collaborate within a cross-functional team of scientists, engineers, product managers, and researchers.
- Publish research findings, deliver talks, and attend top industry conferences.
View Full Description & ApplyYou'll be redirected to the employer's site
