Licenses and sources
Sources and license terms for the voice assets used to create audio.
Pocket TTS Czech
Czech audio uses Pocket TTS Czech (vvolhejn/pocket-tts-czech). The model is derived from ParCzech4Speech by Vladislav Stankov, Matyáš Kopp, and Ondřej Bojar at the Institute of Formal and Applied Linguistics, Charles University.
The model weights and bundled reference voices are available under CC BY 4.0. The Pocket TTS software is licensed separately under MIT.
CC BY 4.0
The model and reference voices may be used, modified, and deployed commercially. Sharing the model, a reference voice, or a modified version requires creator and source credit, links to the original material and license, and a note describing any changes. The credit must not imply endorsement by the authors or Charles University.
CC BY 4.0 does not automatically apply to every generated recording. Rights in the input text and privacy, publicity, or other personality rights connected with a voice remain separate. A recording must not be presented as an authentic statement by the reference speaker.
Supertonic 3
Supertonic 3 by Supertone Inc. provides its model and voice styles under the OpenRAIL-M License. The accompanying software is available under MIT.