CubeLens (iOS) and Cubed (visionOS) · Last updated August 8, 2026
Both apps narrate every solving step in a choice of synthesized voices. Those voices are built from openly licensed speech corpora, and the credits below are part of how they are licensed.
Alba voice: synthesized from a text-to-speech model trained on the Alba speech corpus by Cassia Valentini-Botinhao and Junichi Yamagishi, University of Edinburgh (https://doi.org/10.7488/ds/2506), used under CC BY 4.0. Changes were made: the corpus was used to train a voice model, and no original recordings are included.
The Alba speech corpus is distributed by the University of Edinburgh under the Creative Commons Attribution 4.0 International license, which permits commercial use and adaptation provided the creators are credited, the license is linked, and changes are indicated. Nothing here implies that the creators or the University of Edinburgh endorse these apps.
Both are synthesized from models trained on public-domain recordings from LibriVox. No attribution is required; it is offered here as a courtesy to the volunteer readers whose recordings made them possible.
Synthesized from a model trained on the OHF-Voice voice-datasets corpus, released under CC0 (public-domain dedication). No attribution is required.
Questions or a correction to any credit on this page: support (at) cubelens.app