Machine Learning Researcher, Audio
Core
Design and train large-scale text-to-speech, speech-to-text, and neural audio codec models for real-time voice agents serving millions of calls.
Role type
Senior IC machine learning researcher (audio)
Builds
Production TTS, ASR, and neural audio codec systems for enterprise voice agents
Domain
Telecommunications / Speech AI / Generative AI
Deliverable
production ML models
Required skills
self-supervised learning, generative modeling, neural audio codec architecture, large-scale model training, inference optimization, experimental design, distributed training, data curation, real-time system constraints
Preferred skills
research publications, open source contributions, telephony background, PhD in ML/AI
Technologies
GPU clusters, quantization, kernel optimization, large-scale speech datasets
Responsibilities
Design and train large-scale TTS models for expressive output; Develop neural audio codec architectures; Build and fine-tune robust ASR systems; Curate and process massive audio datasets; Design staged training curricula; Run rigorous ablation studies and validation experiments
Seniority
Senior, hands-on IC