NADI 2026 TASK 5.1 SLU - INTENT RECOGNITION
LiveAI & MLAR / VR

NADI 2026 TASK 5.1 SLU - INTENT RECOGNITION

Organizer: harounelleuch; 18 submissions; The Intent Recognition subtask evaluates systems on their ability to identify the user’s intended action directly from a spoken utterance ...

harounelleuchOrganizer harounelleuch
Official site

About this hackathon

The Intent Recognition subtask evaluates systems on their ability to identify the user’s intended action directly from a spoken utterance in Tunisian Arabic. Given an input audio recording, participants must assign the corresponding intent label from the predefined set of intents in the SLURP-TN dataset. Submissions are evaluated using classification accuracy and weighted F1-score. For more details you can visit the **official NADI 2026 page**: https://nadi.dlnlp.ai/2026/ The **dataset** is available here: https://huggingface.co/datasets/Elyadata/SLURP-TN and **baselines** are available here: https://github.com/elyadata/SLURP-TN-baselines

Tracks

General Track

Organizer: harounelleuch; 18 submissions; The Intent Recognition subtask evaluates systems on their ability to identify the user’s intended action directly from a spoken utterance in Tunisian Arabic. Given an input audio recording, participants must assign the corresponding intent label from the predefined set of intents in the SLURP-TN dataset. Submissions are evaluated using classification accuracy and weighted F1-score. For more details you can visit the **official NADI 2026 page**: https://nadi.dlnlp.ai/2026/ The **dataset** is available here: https://huggingface.co/datasets/Elyadata/SLURP-TN and **baselines** are available here: https://github.com/elyadata/SLURP-TN-baselines

Prizes

1

Project Prize

Organizer: harounelleuch; 18 submissions; The Intent Recognition subtask evaluates systems on their ability to identify the user’s intended action directly from a spoken utterance in Tunisian Arabic. Given an input audio recording, participants must assign the corresponding intent label from the predefined set of intents in the SLURP-TN dataset. Submissions are evaluated using classification accuracy and weighted F1-score. For more details you can visit the **official NADI 2026 page**: https://nadi.dlnlp.ai/2026/ The **dataset** is available here: https://huggingface.co/datasets/Elyadata/SLURP-TN and **baselines** are available here: https://github.com/elyadata/SLURP-TN-baselines

$1,000

Schedule

  1. Jun 15, 04:00 PM

Tags

#AI#Data#Science#Benchmark#Competition