System and method for dubbing by automatically aligning time axis

Abstract

The invention provides a system and a method for automatically aligning time axis dubbing, and the system comprises a silent video extraction module which is used for removing the audio of an original video; the OCR subtitle extraction module is used for generating a subtitle file in an SRT format; the AI translation module translates the subtitles into a target language; the subtitle erasing module is used for positioning and erasing original subtitles in the video; the TTS speech synthesis module is used for generating a dubbing file; the time axis alignment module is used for dynamically adjusting the speech speed, the video speed and the subtitle time; and the video synthesis module is used for integrating the silent video, dubbing and subtitles. According to the method, OCR subtitle extraction, AI subtitle translation and generative dubbing technologies are combined, and the voice speed of the voice is dynamically adjusted to keep the voice aligned with the time axis of the video and the time axis of the subtitle, so that the quality of the translation play can be improved, the cost can be greatly reduced, the product marketing is accelerated, and great convenience and competitive advantages are brought to content creators and issuers.

Images