MultimodalSensor / Time-seriesVideo
Open X-Embodiment
1M+ real-robot trajectories from 20+ institutions.
1M–10M·Apache-2.0·rosbag
AudioMultimodalText
Habibi
A systematic and standardized benchmark for the multi-dialect Arabic zero-shot TTS task Paper: "Habibi: Laying the Open-Source Foundation of…
A systematic and standardized benchmark for the multi-dialect Arabic zero-shot TTS task Paper: “Habibi: Laying the Open-Source Foundation of Unified-Dialectal Arabic Speech Synthesis”Github: !NOTE Most subsets are licensed under Apache 2.0, except for SAU sourced from the CC-BY-NC-SA-4.0 SADA.
Source: Hugging Face Hub (SWivid/Habibi). Metadata imported from the dataset’s Hub tags.