Senior Machine Learning Scientist (ASR)

About the role

Join Sonos' Voice Control team to design and develop cutting‑edge audio‑based AI components for the company's in‑house voice assistant, focusing on far‑field automatic speech recognition and wake‑word detection.

Key responsibilities

• Design and train state‑of‑the‑art models for ASR and wake‑word detection, targeting specialised domains such as music entities and home automation.
• Define and implement data strategies, including audio augmentation to simulate far‑field conditions.
• Maintain scalable training and evaluation pipelines and contribute to the technical roadmap for the ASR domain.
• Mentor team members on ML best practices and collaborate with cloud backend and embedded engineering teams to optimise model performance across environments.

Required profile

• 8+ years of experience in machine learning research and engineering for voice applications.
• PhD or Master’s degree in computer science or a related technical field, or equivalent experience.
• In‑depth knowledge of speech processing, especially modern ASR architectures such as RNN‑T and Transformers.

Required skills

• Advanced Python programming.
• Proficiency with PyTorch.
• Experience with low‑level compiled languages (Rust, C, C++).
• Hands‑on experience with real‑time, low‑latency streaming ASR and model adaptation techniques.