Google's generated video model Veo3 has a subtitle problem

AI Video & Visuals


It's not easy to get rid of them. Users are forced to play clips (which cost more money), use tools to remove external subtitles, or crop the video to remove subtitles completely.

Josh Woodward, vice president of Google Labs and Gemini, posted on X on June 9 that Google developed a fix to reduce meaningful text. But a month later, users are recording issues on Google Labs' Discord channel, showing how difficult it can be to fix issues with major AI models.

Like its predecessor, VEO 3 can pay for members of Google's subscription tier, starting at $249.99 a month. To generate an 8-second clip, users enter a text prompt describing the scene they want to create on Google's AI Filmmaking Tool Flow, Gemini, or other Google platforms. Each VEO 3 generation costs a minimum of 20 AI credits, and accounts can be refilled at a cost of $25 per 2,500 credits.

Advertising creative director Mona Weiss says it's become expensive to play her scenes to get rid of random captions. “If you're creating a scene with dialogue, up to 40% of that output has impossible subtitles,” she says. “You're burning money trying to get the scene you like, but you can't even use it.”

When Weiss reported the issue to Google Labs through the discrepancy channel, hoping to receive a refund for unnecessary credits, the team directed her to the company's official support team. They offered her a refund for the costs of VEO 3, but not for credit. Weiss refused because accepting means losing access to the model entirely. The Google Labs Discord Support team has told users that speeches can trigger subtitles, and they are aware of the issue and are working on fixing it.



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *