Machine Learning Researcher
Neuphonic · London
Infrastructureonsitesenior
At a glance
Highlights
- Advance state-of-the-art speech synthesis from architecture to production deployment.
- Build ultra-low-latency voice models that run on CPU-constrained devices.
- Explore multimodal speech approaches for natural real-time conversational AI.
What they offer
- Founding team role helping set company culture
- Competitive salary and equity stake
- Private health insurance
- Conference, travel and development budget
Ideal candidate
Research-driven ML scientist with deep curiosity for speech and conversational AI who thrives bridging cutting-edge research and real-world systems in a small founding team.
Requirements
- MSc or PhD in machine learning, speech processing, computer science or related field
- Several years of research or industry experience developing ML models, not entry-level
- Strong experience training and evaluating deep learning models
- Experience in TTS or speech synthesis, or multimodal ML involving audio
- Experience in research-oriented ML environments like academic labs or deep-tech startups
- Familiarity with state-of-the-art speech generation and audio modelling
- Experience reading and implementing recent ML research papers
- hiring process
- Review Hugging Face and GitHub repos before applying, as they will be part of interview discussion.
Lab profile
About Neuphonic
Building ultra-low-latency on-device voice AI for realistic, real-time speech.
What they build
- Ultra-low-latency text-to-speech
- Voice cloning
- Multilingual speech generation
Known for
- Ultra-low-latency inference
- Super-realistic human-like speech
- CPU-efficient generative audio models
- Real-time interaction without cloud dependence
Posted · Last seen