Speech Recognition Engineer

  • Science
  • Remote job

Speech Recognition Engineer

Job description

Keywords: stock option program, remote work, flexible hours, speech recognition, Kaldi, Python, machine learning, language modeling


How do you recognize a successful startup? There’s more demand for our service than our product development can deliver – which is why we need you in our team!


Inscripta’s core solutions are composed of handling masses of audio data to a structured text format, efficiently, precisely and securely. 


Did you know that on a daily basis, thousands of patient stories are dictated in our national health care districts, and that hundreds of people are transcribing these dictations, even around the clock?


Inscripta has developed a dictation and transcription solution which substantially increases the efficiency of the aforementioned processes. A key part of this is our own proprietary ML/AI based Speech Recognition module which is already among the best ones out there. We do however want to develop this system further to be the best possible. Not only for one language but for all possible languages. 


This is why we are on the lookout for new talent to help us make these plans a reality.


As a Speech Recognition Engineer at Inscripta you will:

  • Work together with our expert team to design and implement new features of our Speech Recognition system. 
  • Devise tools and techniques to adapt our speech recognition system to new languages.
  • Create new models for different domains of language.
  • Find ways to integrate speech recognition further with other Natural Language Processing and Machine Learning tools to make the transcriptions process faster and better
  • Help our transcribers to do their job in only 25% of the time they would need without ASR.

Requirements

Requirements

  • Research experience in speech recognition, machine learning or natural language processing

or

  • A computer science or mathematical background and 1-2 years of working experience in machine learning, speech recognition or related fields.


And

  • A Masters' degree in a relevant field or a portfolio of speech recognition/nlp/deep learning projects (e.g. your github profile).
  • Strong drive to automate any part of the model building process
  • Strong background in software engineering; being able to bring your ideas alive using solid code
  • Hunger for keeping in touch with the latest technology and improving yourself


What we consider as an advantage

  • Experience with speech recognition toolkits such as Kaldi
  • Knowledge of modern software development practices such as unit test, version control, and continuous integration
  • Experience with AWS or other cloud computing environments
  • Experience with GPU-based training of machine learning models.

What we can offer

  • Great colleagues in a flat organization.
  • Work in a fast growing startup using state-of-the-art technology
  • Highly challenging, but fun work
  • Competitive salary and stock option plan
  • Flexible hours
  • Ability to grow as a professional and visit suitable conferences and workshops
  • Motivating startup environment at Maria01 (https://maria.io) in Helsinki or support in working remotely from anywhere in Europe.