Language Modelling Research Engineer

Location Germany-Aachen | Germany-Ulm
Job ID
Research & Development - Research Engineering
Pos. Type
Full Time

Company Overview

Wir bei Nuance ermöglichen es den Menschen problemlos mit ihren vernetzten Geräten und der digitalen Welt zu kommunizieren.

Indem wir eine Welt schaffen, in der Menschen auf natürliche und intuitive Art und Weise mit Technik interagieren können, tragen wir dazu bei, dass die Technik dem menschlichen Handeln und Denken immer ähnlicher wird.


Unsere dynamische Technologie verwendet Analysetools und fortschrittlichste Algorithmen, um Gebrauchsgegenstände zum Leben zu erwecken und somit alltägliche Prozesse zu vereinfachen.


Werden Sie Teil unseres Mobile Team… intelligente Systeme für unterwegs. Unsere Leidenschaft gilt der Entwicklung von intelligenten Interfaces, die es Menschen ermöglichen allein mit Hilfe ihrer Stimme mit ihren Autos, Smartphones und anderen Geräten zu kommunizieren.


Job Summary

Summary: Integrate developments of the Corporate Research LM team into the VoCon embedded recognition engine for deployments in real life projects.

Work with the LM teams to organize the production of dictation models in a multitude of languages in service of the divisional business goals.

Reporting structure: You will report to Jan Vuylsteke (group manager Automotive Languages R&D).



  • Interact with the Corporate Research team on language modeling and text formatting technologies for multiple languages. Leverage these technologies in the divisional research and product development programs, targeted at the embedded automotive space • Work with colleague researchers and engineering teams to bring the best LM technology solutions to Nuance's customers of the embedded product lines • Advance the state of the art of automotive embedded dictation by developing innovative systems towards real life deployments, and participate in product development programs, with the team working on the speech engine FST decoder • Productization of software, turning research code into stable production software, optimized for embedded platforms



Number of Years of Work Experience: 2+


Required Skills:

  • Experience in language modeling for speech dictation systems (essential) • Good knowledge of speech recognition systems *(nice to have) • Knowledgeable about the state-of-the-art in language modeling such as neural networks and deep learning based methods • Programming skills in Python and C/C++ (desirable) • Good user knowledge of Linux and Windows systems • Excellent oral and written communications skills in English. • Ability to carry out focused and goal-oriented research and development, ability to assume responsibility for one’s work • Ability to work in an international, cross-site team. Being a team player • Ability to creatively develop innovative ideas while leveraging existing technology with an eye for efficiency.


Preferred Skills:

  • Good Knowledge of statistics, pattern recognition, information retrieval • Expertise in FST based or other ASR decoding techniques is a big asset • Understanding of methods to resolve language specific challenges of dictation and text formatting tasks • Operational knowledge of languages other than English



  • Master degree in physics, mathematics, computational linguistics or engineering • PhD or equivalent industrial experience in a relevant field is a plus


This role can be based in Belgium or Ulm, Germany.


We would also consider recent graduates.



Sorry the Share function is not working properly at this moment. Please refresh the page and try again later.
Share on your newsfeed

Connect With Us!

Not ready to apply? Connect with us for general consideration.