Create and release your Profile on Zintellect – Postdoctoral applicants must create an account and complete a profile in the on-line application system.  Please note: your resume/CV may not exceed 2 pages.

Complete your application – Enter the rest of the information required for the IC Postdoc Program Research Opportunity. The application itself contains detailed instructions for each one of these components: availability, citizenship, transcripts, dissertation abstract, publication and presentation plan, and information about your Research Advisor co-applicant.

Additional information about the IC Postdoctoral Research Fellowship Program is available on the program website located at: https://orise.orau.gov/icpostdoc/index.html.

If you have questions, send an email to ICPostdoc@orau.org.  Please include the reference code for this opportunity in your email. 

Research Topic Description, including Problem Statement:

Recent deep learning developments have produced dramatic improvements in the performance of speech recognition technologies. Many advancements are due to end-to-end (E2E) neural network models, which optimize a direct mapping from audio data directly to text information with a single network. However, the supervised manner in which these E2E models are trained does not produce state-of-the-art performance results in adverse conditions, such as noisy audio environments and low-resource language scenarios.

With this project, we aim to identify methods to improve E2E speech recognition performance robustness for noisy audio environments and low-resource language scenarios.

Example Approaches:

• Advanced speech enhancement modules could be developed as a data preprocessing step for noisy audio environments. Although using speech enhancement modules to improve the listening quality of noisy audio is a well-established research technique, it is known to be harmful as a speech recognition front-end module. Solving this degradation issue is an active research challenge.
• Pre-training/fine-tuning approaches could be explored for low-resource language scenarios. Supervised E2E models require large amounts of labeled data to achieve state-of-the-art performance results, which is an obstacle for achieving comparable results for low-resource languages. Exploring approaches using self-supervised learning representation models pre-trained with multiple languages and fine-tuning for low-resource target languages is an active research area that could be explored. 

Relevance to the Intelligence Community (IC):

Speech recognition is a fundamental Human Language Technology (HLT) capability that provides audio content triage capabilities. New state-of-the-art E2E neural network approaches remove onerous large, labeled data requirements needed to develop a speech recognition capability for a language of interest; however, these gains are at the expense of lower performance results in adverse audio environments and low-resource scenarios. The work described above would identify solutions to these issues which are critical to maintaining a competitive edge in language analysis capabilities.

Postdoc Eligibility

• U.S. citizens only
• Ph.D. in a relevant field must be completed before beginning the appointment and within five years of the application deadline
• Proposal must be associated with an accredited U.S. university, college, or U.S. government laboratory
• Eligible candidates may only receive one award from the IC Postdoctoral Research Fellowship Program

Research Advisor Eligibility

• Must be an employee of an accredited U.S. university, college or U.S. government laboratory
• Are not required to be U.S. citizens

Key Words: #Speech Recognition, #Multilingual, #Deep Learning, #Neural Network, #End-to-end, #Low Resource, #Noise Robustness

• Citizenship: U.S. Citizen Only
• Degree: Doctoral Degree.
• Discipline(s):
  • Chemistry and Materials Sciences (12 )
  • Communications and Graphics Design (5 )
  • Computer, Information, and Data Sciences (17 )
  • Earth and Geosciences (21 )
  • Engineering (27 )
  • Environmental and Marine Sciences (14 )
  • Life Health and Medical Sciences (48 )
  • Mathematics and Statistics (11 )
  • Other Non-Science & Engineering (2 )
  • Physics (16 )
  • Science & Engineering-related (1 )
  • Social and Behavioral Sciences (29 )