Skip to content

synthere/syndub

Repository files navigation

SynthereDub: Video Dubbing tool that support using reference audio


[English](README.md)

mainframe SynthereDub Translate videos from one language's voice to another using free speech recognition and translation tools, with built-in edge TTS speech synthesis, and can also use voice synthesized from third-party tools.

IMAGE ALT TEXT HERE

Updates

  • 2024-10-30 Support the use of reference audio generated by third-party TTS.
  • 2024-11-22 Support multi-speaker video, currently 2speakers, for which different TTS voice can be set.

Highlights

Synthere Dub has the following features:

  1. Adaptive alignment Text and speed challenges to achieve precise alignment with the original video sound.
  2. No need to register for third-party tools Use offline speech recognition Whisper, integrated with the free online speech synthesis tool edgeTTS.
  3. Support third-party sound conversion Supports sound from other sources as the destination sound.
  4. Multiple speaker diarization Supports dubbing more a video with more than 1speakers.

Hardware requirements

Can run on CPU or GPU. The CPU runs slower when running the advanced version. If running on a GPU, you need at least 8GB of video memory; It is recommended to use 16GB or more of memory for CPU operation.

##Supported Languages Currently, only English to Chinese translation is supported.

Download and use

References

Mainly relies on the following open-source projects

About

Video dubbing with tts or other audio

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Sponsor this project

Packages

No packages published

Languages