Dataset Groups Activity Stream BABEL The BABEL dataset is a multilingual speech recognition dataset containing over 1,000 hours of speech from 6 languages. BibTex: @dataset{Haodong_Duan_and_Jiaqi_Wang_and_Kai_Chen_and_Dahua_Lin_2024, abstract = {The BABEL dataset is a multilingual speech recognition dataset containing over 1,000 hours of speech from 6 languages.}, author = {Haodong Duan and Jiaqi Wang and Kai Chen and Dahua Lin}, doi = {10.57702/3tt2d3qo}, institution = {No Organization}, keyword = {'25-joint skeleton', 'Dataset', 'Multilingual', 'Speech Recognition', 'action analysis', 'action-driven', 'body behavior', 'human motion', 'human movement', 'prediction'}, month = {dec}, publisher = {TIB}, title = {BABEL}, url = {https://service.tib.eu/ldmservice/dataset/babel}, year = {2024} }
