Google Ads video voice-over: 5 messages, 150 characters each

Google Ads voice-over lets you add up to 5 voice-over messages to a video, each up to 150 characters, in a language and voice you select, with audio durations of 6 to 15 seconds, 15 to 30 seconds, or 6 to 30 seconds. That is enough for a short localized call to action, not for a translated narration of a talking ad. For longer scripts or a spoken avatar, Sume's avatar video takes scripts estimated at 4 to 60 seconds.

The Google facts come from Add voice-over to a Video campaign, read on 2026-10-02. The Sume facts come from docs.sume.com.

What are the voice-over limits?

The page calls it a free way to add text-to-speech voice-over to video campaigns through the Asset Library. You choose a language, a voice, and a volume (default 100%). You set the start time of each message, and messages cannot overlap.

The page does not list the available languages or voices, and it does not mention dubbing or whether the feature replaces existing audio.

What does that mean for a localized ad?

Five messages of 150 characters is at most 750 characters in total, and 5 times 150 only holds if every message uses its full length. For a spoken hook, an offer line, and a call to action in each market, that is workable: write each message in the target language, keep it under 150 characters, and set start times so none overlap.

A translated script usually reads longer than the English source, so count characters after translation, not before. A line that fit at 140 characters in English can pass 150 in another language.

What if the script is longer than 150 characters?

Then you need a different route for the voice. Sume's avatar video accepts a script or video_inputs when Sume estimates the video at 4 to 60 seconds, and says to shorten or split longer ones into multiple jobs. That produces a talking avatar clip, not a voice-over laid on your existing footage.

If you already have audio from your own text-to-speech or a recording, timeline audio can concat up to 20 Sume-hosted audio parts into one gapless file, with no re-synthesis, for $0.01 per job. That joins audio; it does not translate or voice text.

How do you add captions in the same language?

Voice-over alone leaves viewers on mute with nothing to read. Video captions can burn the same message as cues with text, start, and end, at $0.20 for video up to 60 seconds per language. That gives a muted viewer the text and keeps the wording identical to the spoken line.

Keep the times of the cues equal to the start times of the voice-over messages in Google Ads so the words and the voice land together.