onsite
NLP Engineer - Speech - 42dot
Research Engineer
Join a cutting‑edge mobility AI team as an NLP Engineer focused on speech. Develop rule‑based and model‑based sentence splitting and text normalization models, optimize them for server and on‑device use, and enhance multilingual text pipelines for superior voice service quality.
About the role
Key Responsibilities
- Design and implement rule‑based and model‑based sentence splitting and text normalization (TN) models for speech applications.
- Optimize TN models for both server‑side and on‑device deployment, ensuring low latency and high accuracy.
- Generate, curate, and maintain high‑quality training data for sentence splitting and TN tasks.
- Develop multilingual text preprocessing pipelines and handle language‑specific nuances.
- Implement post‑processing and exception handling to improve overall voice service quality.
Requirements
- Minimum 2 years of experience in NLP or related fields.
- Strong understanding of language processing fundamentals and NLP concepts.
- Experience with DNN‑based tokenizers and text processing frameworks.
- Proficiency in Python and familiarity with machine learning libraries (e.g., PyTorch, TensorFlow).
- Ability to work in a fast‑paced, collaborative environment.