Open Issues Need Help
View All on GitHub 实现ASR以及ASR+OCR结果融合 2 months ago
AI Summary: Implement ASR functionality to transcribe audio segments detected by VAD, and integrate a large language model to fuse ASR and OCR results for improved accuracy in video transcription. Currently, only OCR-based transcription from hard subtitles is implemented.
Complexity:
4/5
help wanted
一个用于从视频中提取语音转录的开源工具，通过融合OCR（硬字幕识别）和ASR（语音识别）技术，提供高质量的转录结果。
Python