LRE i-vector participants may find the key for the dev segments under Donwload/Datasets

NIST LRE i-vector challenge

The National Institute of Standards and Technology (NIST) LRE i-vector challenge

The National Institute of Standards and Technology (NIST) will coordinate a special i-vector challenge in 2015 based on data used in previous NIST Language Recognition Evaluations (LREs). This challenge is intended to foster interest in this field from the broader machine learning community.
It will be based on the i-vector paradigm widely used by state-of-the-art speaker and language recognition systems, and will largely follow the paradigm of the recent NIST coordinated Speaker Recognition i-Vector Challenge . By providing i-vectors directly, and not utilizing audio data, the evaluation is intended to be readily accessible to participants from outside the audio processing field.
This challenge is designed to focus on the development of new methods for using i-vectors for language identification in the context of conversational telephone speech. It is designed to foster research progress, including goals of

  • Exploring new ideas in machine learning for use in language recognition
  • Making the language recognition field accessible to more participants from the machine learning community
  • Improving the performance of language recognition technology
To learn more about the challenge, please read this evaluation plan or send questions to ivector_poc@nist.gov.



Register to evaluation