Update README.md

README.md

@@ -9,13 +9,13 @@ dataset_info:
     dtype: string
   splits:
   - name: train
-    num_bytes: 49701748.0
+    num_bytes: 49701748
     num_examples: 1104
   - name: validation
-    num_bytes: 8575145.0
+    num_bytes: 8575145
     num_examples: 179
   download_size: 58269230
-  dataset_size: 58276893.0
+  dataset_size: 58276893
 configs:
 - config_name: default
   data_files:
@@ -23,4 +23,28 @@ configs:
     path: data/train-*
   - split: validation
     path: data/validation-*
+task_categories:
+- automatic-speech-recognition
+language:
+- br
 ---
+
+
+# Description
+Partie en breton du jeu de données [ML-SUPERB 1.0](https://multilingual.superbbenchmark.org/).  
+En pratique nous nous sommes basés sur [espnet/ml_superb_hf](https://huggingface.co/datasets/espnet/ml_superb_hf).  
+Selon nos estimations, ce jeu de données représente 1 h 10 min et 7s.
+
+
+# Citation
+```
+@misc{shi2025mlsuperbmultilingualspeechuniversal,
+      title={ML-SUPERB: Multilingual Speech Universal PERformance Benchmark}, 
+      author={Jiatong Shi and Dan Berrebbi and William Chen and Ho-Lam Chung and En-Pei Hu and Wei Ping Huang and Xuankai Chang and Shang-Wen Li and Abdelrahman Mohamed and Hung-yi Lee and Shinji Watanabe},
+      year={2025},
+      eprint={2305.10615},
+      archivePrefix={arXiv},
+      primaryClass={cs.SD},
+      url={https://arxiv.org/abs/2305.10615}, 
+}
+```