The 2017 language recognition evaluation (LRE17) is the next in an ongoing series of language recognition evaluations conducted by NIST since 1996.
The objective of these evaluations are:
- to support language recognition research by exploring promising new ideas in language recognition and developing advanced technology incorporating these ideas
- to measure and calibrate the performance of language recognition systems
They are intended to be of interest to all researchers working on the general problem of text- and speaker-independent language recognition. To this end, the evaluation is designed to be simple, to focus on core technology issues, to be fully supported, and to be accessible to those wishing to participate.
The basic task in NIST’s language recognition evaluations is language detection, i.e., given a segment of speech and a target language, automatically determine whether the target language was spoken in the test segment. The 2017 evaluation is similar in many ways to LRE15, but in addition to conversational telephone speech (CTS), speech excerpts extracted from online videos will be used as development and test material. For more information about LRE17 please read the evaluation plan
or send questions to firstname.lastname@example.org
NIST / IAD / MIG