Requirements

• PhD in computer science, statistics, applied mathematics, or a related field
• Peer-reviewed publications at venues such as ICASSP, ISMIR, Interspeech, NeurIPS, ICLR, ICML, or related
• Expertise in generative modelling, machine learning, music information retrieval, or speech and audio processing
• Industry experience coding in Python and PyTorch or JAX
• Excellent communication and presentation skills

Nice to Haves

• Deep expertise in audio generation and editing, post-training of diffusion models, or voice synthesis
• Experience integrating machine learning models into consumer-facing applications
• Love of music

What You'll Be Doing

• Conduct research in generative audio using diffusion or flow-matching models
• Focus on voice generation, vocal and speech synthesis, and related audio and signal processing techniques
• Develop post-training and alignment techniques to improve output quality, controllability, and adherence to human preference
• Create architectures and distillation methods for realtime generation with low latency
• Build evaluation methodology and benchmarking systems for state-of-the-art comparison
• Run large-scale experiments on internal infrastructure
• Collaborate with engineering, product, and design teams to transition research into real listening experiences
• Stay connected with the broader research community