Inworld SAVA Copyright 2026 Inworld AI The model weights were trained on Inworld first-party data, including data purchased from vendors under private terms, and on these third-party datasets: - pipecat smart-turn data v3.2, train split: CC BY 4.0, https://huggingface.co/datasets/pipecat-ai/smart-turn-data-v3.2-train - VoxConverse, dev split: CC BY 4.0, https://github.com/joonson/voxconverse - VocalSound: CC BY-SA 4.0, https://github.com/YuanGongND/vocalsound - Tatoeba (CC BY 2.0 FR, https://tatoeba.org), FineWeb-2 (ODC-BY 1.0, https://huggingface.co/datasets/HuggingFaceFW/fineweb-2) and C4 (ODC-BY 1.0, https://huggingface.co/datasets/allenai/c4), as text for synthetic speech