WEBVTT

00:00:00.000 --> 00:00:03.002
This example uses a synthetic voice. Use a real speaker’s voice only with their permission.

00:00:03.002 --> 00:00:11.833
1. Select the spoken passage. Open Speakers and wording → Change spoken words…

00:00:11.833 --> 00:00:24.119
2. Edit the line. Check the voice reference and confirm permission to use that voice.

00:00:24.119 --> 00:00:28.517
3. Generate voice preview. Generation runs locally on the CPU.

00:00:28.517 --> 00:00:30.519
Processing locally. This waiting time is shortened in the tutorial.

00:00:30.519 --> 00:00:32.520
Results ready. About 5 seconds of processing time were shortened.

00:00:32.520 --> 00:00:41.870
Listen to the preview before accepting. The passage keeps its length and edit fades.

00:00:41.870 --> 00:00:49.409
4. Accept audio replacement keeps the preview. Ctrl + Z restores the original audio and words.

00:00:49.409 --> 00:00:53.374
Keep only a preview you are happy with. Correct word remains the simpler text-only correction tool.