Samsung TV Voices does not include Samsung firmware, speech engines, models,
or voices. When requested by the user, it downloads a supported package from
Samsung's official download service, verifies its fixed SHA-256 digest, keeps
only the speech components in the NVDA user-data folder, and removes the
firmware and extraction workspace.

Firmware package separation is performed by unixtract:
https://github.com/theubusu/unixtract
License: GNU General Public License version 3

VDFS filesystem extraction is based on Samsung's vdfs4-tools.0010 source,
distributed through the HinTak/vdfs-tools repository:
https://github.com/HinTak/vdfs-tools
License: GNU General Public License version 2 or later

The local VDFS modification skips all regular files except the Samsung speech
tree and speech-engine library. Complete corresponding source, including that
change and build instructions, is provided in the public source repository for
this add-on.
