Skip to content

Dataset · FBK-MT

Test split of the multilingual SLU dataset

FBK-MT/Speech-MASSIVE-test

Test split of a multilingual spoken corpus for speech recognition and intent understanding.

What's inside

The repository contains only the test split (10 to 100 thousand records) of the spoken counterpart to the text MASSIVE corpus, while train and dev splits are hosted separately. It includes audio recordings and texts in multiple languages, including Arabic, German, Spanish, French, Hungarian, and Korean (12 languages in total in the project).

What you can do with it

  • Evaluating ASR model quality across multiple languages
  • Testing spoken language understanding (SLU) and intent classification systems
  • Benchmarking zero-shot classification on voice commands
Tasks
Audio classification, Text classification, Zero shot classification, Automatic speech recognition
Languages
Arabic, German, Spanish, French, Hungarian, Korean, Dutch, Polish and 4 more
Size
10K–100K rows
Downloads this month
377
Likes
11
Access
on request on Hugging Face

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("FBK-MT/Speech-MASSIVE-test")