The language of the narration. It switches the text and the voices at once, and speaks the text straight away. The three texts voiced by a person were recorded in English, so in Dutch every text is generated.
American, male
British, male
American, female
British, female
Cinematic suspense, theatrical and precise. The heaviest choice for a dramatic announcement.
Choosing a voice speaks the text straight away. The marked voices are default ElevenLabs voices and expire at the end of 2026, so for the stage a library voice is the safer choice.
Reads directions in the text itself, such as [calm] or [pause], and shapes the delivery around them. Slightly quicker paced than v2 on the same words.
Choosing a model speaks the text straight away, so the two are one click apart. Only v3 hears the directions in square brackets; the others speak the text as it stands.
A marked text is ready on the server and starts straight away. The rest is generated on the first play, so the marks follow the language, the voice and the model you pick.
edit the text
the DC26 set
separate sound for tuning
Vapour · arrows switch · v compare · f fullscreen · h ui · d stats · a audio
save this look
saved in this browser
Nothing saved yet. Whatever you save here stays in this browser.