We are looking for a Product Linguist to develop and tune speech recognition systems for
deployment in consumer products. As this position will require working with multiple languages
outside of English, a strong knowledge of general linguistic theories is required. The successful
candidate will have a demonstrated background in phonology and phonetics and/or automated
speech recognition or machine learning. The candidate will also have a record of working
independently with strong time management skills, flexibility, and reliability.
DUTIES AND RESPONSIBILITIES
Develop, review, test and improve linguistic components of speech recognition and synthesis
- Design data collection protocol and evaluate and clean speech data for machine
- Participate in detailed investigations of voice recognition performance for
customer and internal projects.
- Learn to use internal tools while maintaining expertise of Sensory technology
- Interact with customers to provide guidance on dialogue design, user interface,
and recognition set makeup for optimal speech recognition success.
- Collaborate with software engineers, applied research scientists, computational
linguists, and product managers on complex solutions for speech recognition
PRIMARY SKILLS / SPECIALITY / TECHNICAL SKILLS
- Strong background in general linguistic theory
- Experience or fluency with multiple languages
- Experience with audio tools such as Praat, SoundForge, Audacity, or Audition
- Ability to interpret and analyze a spectrogram.
- Strong command of phonetic transcription such as IPA, SAMPA, or WorldBet (preferred)
- Familiarity with command line functions and Unix/Linux.
- 3+ years experience in computational linguistics or related field
- BA or higher in Linguistics or related fields such as Computer or Data Science with a
focus on language.
- Professional experience scripting with one or more computer languages such as Python,
Bash, TCL, or other.
- Excellent understanding of phonetics, phonological, and acoustic concepts
- Comfortable with version control systems, preferably Git
- Demonstrated ability in manipulation of large data sets, including transcription,
annotation, and QA
- MS degree in Linguistics, Computational Linguistics, Data Science/Computer Science
with linguistics interest, or related field
- Proficiency in at least one language other than English (Korean, Japanese, Portuguese,
Spanish, French, German preferred)
Job Code# 2301-GH