Developer Seeks Fix for Gibberish in Long-Take Voice Cloning with WAV Samples

davekilljoy · reddit · 2026-08-09

A developer is struggling to achieve consistent voice cloning using a 10-second WAV reference. When applying this to 10-15 second long single-camera shots, the character outputs gibberish 90% of the time.

Currently, the only workaround is using multiple cuts in the video. The author is asking the community for a reliable way to generate cloned audio for long-running clips without relying on frequent cuts.

Original post →

More from Multimodal

Multimodal channel →