Pika, a rival of OpenAI, has unveiled a new feature for its paying subscribers called Lip Sync. This feature allows users to incorporate spoken dialog into videos using AI-generated voices from ElevenLabs, a generative audio startup. The voices are synchronized with animated characters to ensure realistic lip movements. Available in “early access” to Pika Pro users and select members of the “Super Collaborators” program, this feature offers text-to-audio and uploaded audio track support, giving users flexibility in voice styles and dialog input methods.
While Pika’s AI video quality may not match competitors like OpenAI’s Sora or Runway, the introduction of Lip Sync sets it apart by addressing a key barrier in creating narrative films with AI. Most other leading AI video generators lack a similar feature, forcing users to rely on third-party tools for adding spoken dialog and lip movements, resulting in lower-quality videos. Notably, Runway has also upgraded its Multi Motion Brush feature, now adding region detection to simplify object selection for motion effects without manual input.
Ed Newton-Rex, CEO of Fairly Trained, expressed concerns about how Pika’s video model was trained, highlighting the importance of transparency and consent in using AI technologies. Despite these inquiries, the competition among video AI generator companies continues to intensify, driving innovation and higher-quality outputs. This rapid evolution raises concerns within the filmmaking community, exemplified by Tyler Perry’s decision to halt a production studio expansion after witnessing the capabilities of Sora-generated videos. As the technological “arms race” in video AI progresses, professionals in the industry remain vigilant about potential impacts on traditional filmmaking practices and employment.
HEALTH
WORLD AFFAIRS
TECHNOLOGY
CREATIVE
