Emploi [CDI] : TAL

[Offre diffusée le 9 novembre sur PCrecruiter]

Speech Scientist, Self-Service Solutions

Location: Northern New Jersey or work remotely, depending on candidate


The company’s solutions are used by major Communications Services Providers globally. Products empower FAST, EASY, and USEFUL automated self-service solutions across all the major contact channels for our customers’ subscribers. These contact channels include traditional voice calls, smart phones and tablets using native apps as well as mobile web, PCs using desktop web, STBs/smart TVs, and other emerging ‘smart’ devices.

As a Speech Scientist, will make key contributions to a suite of pre-built applications that interact with callers/users intelligently over their devices using advanced Natural Language Understanding and Speech Recognition technology to provide advanced self-service capabilities ranging from natural language call routing to technical support to billing functions and more.

These solutions are built using a suite of integrated software development tools and are fully compliant with industry leading and de-facto standards for speech applications. Products are fully deployed and operational in our own data-centers and fully managed, operated, and maintained by the company, using top-tier service providers to support over 15 million subscribers.


Overall responsibility for leading the company’s Speech Science Practice to maintain and improve the speech recognition technology, performance, and accuracy of NLU and speech recognition technology. Will create speech-enabled solutions and functionalities for new and existing solutions.

  • Work closely with Product/Project Managers, Architects, Business Analysts, VUI Designers, Software/Systems Engineers, Customers, and Partners to understand and define the product roadmap and requirements from a speech recognition perspective
  • End-to-end grammar development and performance improvements of new and existing voice solutions
  • Streamlining existing grammars to maximize and promote reusability and common code
  • Innovating and improving continuous application grammar tuning process – especially around Natural Language Grammars and Statistical Language Models
  • Maintaining, supporting, enhancing, and improving grammars
  • Learning to use all existing Speech Science tools, and then helping to maintain and improve these tools
  • As a key member of a cross-functional team, will monitor, analyze, and tune speech-enabled products on a regular basis to improve performance against three key dimensions:
  • Application effectiveness
    • how well is the application performing against the business objectives, for example, maximizing completion-rate
  • Caller/User experience
    • overall experience in using the application
  • Speech recognition accuracy
    • how well does the application understand what the callers are saying
  • Identifying, defining, improving, and applying best-practices in grammar design, development, and tuning
  • Delivering high-quality application/grammar releases to QA and into production with zero or minimal defects
  • Working with all cross-functional team members both within and outside the company <x{(}>(e.g. customers, partners, and vendors
  • Staying current on the latest industry trends, technologies, practices, and innovation in speech recognition technology – introducing relevant ideas into the company
  • Representing the company as a Speech Science Subject Matter Expert in customer facing meetings, conferences, trade-shows, and seminars
  • Promoting the company’s brand and presence in the Speech Science community by publishing technical papers and articles


Job Requirements:

    • Minimum: A Bachelors degree in Computer Science, Machine Learning, Linguistics, or a related field
    • Preferred: Masters or PhD
    • 4 to 8 years of direct experience in speech science research and development
      • Experience in developing large-scale complex grammars with very large vocabulary and/or statistical language models
      • Experience working with industry-leading speech recognition engines (e.g. Nuance, AT&T, Microsoft, etc.)
      • Experience working with industry-leading VoiceXML browsers (e.g. Genesys, Avaya, Nortel, etc.)
    • Excellent communications skills – written and oral
    • Willingness to travel as required
    • Desirable:
      • experience in the Communications Services industry
    • published author in speech science (books, papers, journals, etc.)