字幕生成器
.mp4 视频的字幕生成脚本。
该脚本将 mp4 文件作为输入,并能够根据音频的静音间隔生成时间戳。 根据音频质量,可能需要更改参数。
THRESHOLD = 90 --> Value of threshold to measure silence
MAJORITY = 0.6 --> Silence is determined based on a majority vote over a small time interval. Depending on the quality of audio, this value needs to be changed. I currently set it to 60%
使用了额外的库和 API
FFMPEG
Google Speech API
随意提出更改建议
2022-04-06 15:30:58
3KB
Python
1