Text-to-Speech Voiceovers
Choose a speech engine, language, and voice to create a previewable narration track.
Open Tools → Text to Speech. Enter the script, choose a voice engine, language, and voice, then select Generate Audio and listen to the preview.
The local engines are Kokoro, Supertonic, and Hojo TTS Light 40M. Models or voices download on first use and are cached locally; Hojo's first download is about 252 MB. Scribis appears as an engine when the Scribis plugin is enabled and loads its available voices from Scribis.
When the audio is ready, select Save to Library, name the WAV, and save it. If you opened TTS from a caption clip, you can instead add the generated audio at the caption's start. Review the wording, language, and voice before adding the audio to the finished mix.
Ready to start?
Apply what you just learned directly in Video Weaver.