Back to all jobs

Applied Scientist – Speech Recognition

Wexford, PA 15090
Lucas Systems, Inc.
ago
fulltime ASR TTS LM NLP

As the Applied Scientist – Speech Recognition at Lucas Systems, you are at the forefront of innovation. Always challenging the status quo, your team is discovering new techniques to capture and transform data to inform, optimize and disrupt a rapidly changing and expanding industry. Our cross-functional team of experts know the only constant is change…and that excites them!

As a member of Lucas’ Machine Learning and Optimization team, you will work in collaboration with some of the brightest, most talented software engineers in Pittsburgh. You’ll enjoy knowing that your team is making a recognizable impact to the growth of a profitable, established but entrepreneurial, software solutions company. Your primary focus will be in building cutting-edge analytics and data science solutions, performing statistical analysis, and building high quality prediction systems integrated with our suite of products. You’ll help build applications that can be productized for the market as part of the entire Lucas platform and experience.

The Applied Scientist – Speech Recognition will be responsible for:

  • Working with 3rd party speech recognition vendor to optimize performance of our speech platform
  • Staying up to date on speech recognition research to ensure the Lucas Speech Platform offers the best customer experience possible
  • Applying data science techniques and digging into weaknesses to analyze system performance of our current ASR models, in order to identify further areas for improvement
  • Working with product development team members to bring to market high integrity, analytic solutions that align with organizational and revenue goals
  • Applying a range of NLP methods in practical application, as well as developing NLP components: writing rules, building grammars/FSTs, creating annotation systems or mappings for the target language
  • Being the resident advisor to cross-functional teams with regard to analysis, insights and statistical methodologies. Clearly articulating the same to a wider audience as well (e.g., presenting at trade shows, writing white papers and blogs, etc.)

Qualifications:

  • MS or PhD in Computational Linguistics, Electrical Engineering, Computer Science, or equivalent in a related discipline
  • 5+ years of experience using Python, C/C++, Java, and SQL or other programming languages
  • 2+ years of experience working with speech recognition toolkits (such as HTK, Kaldi, SRILM, OpenFST, or equivalent proprietary systems)
  • Strong knowledge of ASR concepts and research trends
  • Bias toward action. Do what it takes to ramp up and solve problems quickly. Ability to synthesize large amounts of data to make informed decisions and act on them.
  • Excellent verbal and written communication skills. Ability to clearly and concisely articulate complex ideas to various groups.
  • Demonstrated commitment to producing high quality, well-designed and flexible solutions
  • Speech recognition and speech synthesis (TTS) research or development experience (speech recognition accuracy evaluations, use of APIs and/or tuning of speech recognition systems like CMU SPHINX and Nuance VoCon 3200, etc.)

Apply Now