Voiceover, music and captions are assembled in the same project as the scenes: the voice is created in Generation, and the tracks are mixed on the Editor timeline.
Voiceover from text
In the Voiceover mode of Generation, enter the text, pick a model and a Voice from the base set or your own.
The finished voiceover goes on the Voiceover track in the Editor, where you attach it to a scene or place it anywhere in the video and set its Volume in decibels separately from video sound and music.
| Model | Characters per request | Delivery control |
|---|---|---|
| Eleven v4, Eleven v4 Turbo | Up to 2,000 | Emotion tags in square brackets such as [whispers] or [laughs], and the Expressiveness and Similarity settings |
| Eleven v3 | Up to 4,000 | The same tags and a Delivery: Natural, Calm, Expressive or Ad |
Voice replacement, your own voice and the glossary
If a recording already exists, Eleven v2 replaces the voice in it while keeping the speech, pace and intonation. Pick the Voiceover source: Project audio/video, Audio or video file, or Microphone, up to 5 minutes long.
- Your own voice is created in Library → Voices → My voices → New voice: add 1–2 minutes of clean speech as audio, video or a microphone recording. Voice training requires an active plan, and the finished voice appears in the Voice picker next to the base voices.
- Words a voice gets wrong are fixed in Voices → Glossary: fill in “Word or phrase” and “Pronounce as”, then check it with Preview. A rule applies to every voiceover generated from text.
Captions
Captions are built in the Subtitles panel. A voiceover already carries its text and word timings; for a video with sound, press Transcribe and the transcription cost is shown on the button before it runs.
Each cue can be edited, split, locked or deleted, and the whole track gets a style (Broadcast, Clean pop or Soft karaoke) and one of nine on-screen positions. Captions keep the offer and key phrases on screen when the sound is off; the voice carries meaning and pace.
Dubbing an existing video
- Sync Lipsync v3Syncs a speaker's lips in a finished video or a photo to the chosen voiceover (the sync-3 model). Works both for replacing speech in filmed footage and for a new talking video.8 tk/ s
- H3 Max Lip SyncAnimates one photo to a ready voiceover: a talking clip as long as the audio, up to 14.8 seconds, from 480P to 2K. It starts from a photo and does not dub filmed video.from3 tk/ s
- Kling Lip-SyncRe-times lip movement in a finished video to a new audio track, for dubbing, localization and replacing lines. It does not generate video.3 tk/ s
When it is useful
Explainer videos
Use voice for flow and captions for the key points.
Ad creative
Check that the offer, product shot, and final call to action fit together.
Client drafts
Review voice, captions, and scene rhythm in one place before approval.
FAQ
When should voiceover be added?
Add it after the first visual structure is clear, but before final export. Audio often changes timing.
Can I choose the voice?
Yes. The Voiceover mode has a Voice picker with base voices and your own. In Eleven v4 the delivery comes from tags in the text and the Expressiveness and Similarity settings; in Eleven v3, from tags and a Delivery: Natural, Calm, Expressive or Ad. Every voice can be previewed before generation.
Why add captions if there is voice?
Captions keep the message and the offer on screen when the video plays without sound, for example in Reels, TikTok and Shorts.
What should be checked before publishing?
Check readability, timing, scene rhythm, object coverage, and whether the video works with and without sound.
