Contribution terms
- I record my own voice, or sounds in which nobody else can be heard in a private conversation. TV, radio, music, street noise and people who agreed to be recorded are fine.
- Background recordings: I record only my own surroundings. I don't record other people's private conversations without their permission, because that is illegal.
- Files I upload are recordings I made myself.
- I grant the project a free, non-exclusive licence to use my clips to train and test the Trstenko wake-word model, for as long as my consent stands.
- I don't say or type names, addresses or other personal details in clips or in the note.
- The project may reject or delete any clip without giving a reason.
- Once your clips are recorded and confirmed (approved), they become part of the training data. Training takes a lot of work, so a model that was already trained on your clips is kept as it is. It is not deleted and not retrained without them.
You can withdraw at any time on the My data page. Withdrawing deletes your audio files and stops any future use of them, but it cannot undo training that has already happened (see part 9 of the privacy notice).