autosubtitle / auto split for audio.
feature
- auto generating subtitle(srt format ) or text for audio data
- cut audio data(wav format ,1 channel) to small part by speakers pause
notice
- only support 1 channel wav file
- user need to retrieve audio data from videos
- generate subtitle/text for video data
- the recognize rate depend on many factors: the qulity of the video data etc
- please apply Baidu api key for using ,contact if you have any question
- ted80001.wav is generated from https://ia800204.us.archive.org/25/items/AomawaShields_2015U/AomawaShields_2015U.mp4
- ted80001.srt is auto generated by ted80001.wav
语音学习
功能
- 自动生成语音字幕
- 可以根据说话人的停顿,进行自动片段剪辑
注意事项
- 只支持 1 个通道的 wav 文件
- 如果进行视频字幕自动生成,用户需要自己提取一个通道的wav文件
- 语音文件识别的字幕格式srt
- 也可以进行语音转化成文字
- 识别率还可以,依赖音频文件的噪声,演讲,朗读音频较好
- 底层使用的Baidu的语音识别,如果使用请自行申请,如果有问题可以联系我
- ted80001.wav 来源于视频 https://ia800204.us.archive.org/25/items/AomawaShields_2015U/AomawaShields_2015U.mp4
- ted80001.srt 由ted80001.wav 自动生成
仅供参考##
知乎上详细的说明 https://zhuanlan.zhihu.com/p/28347508
音乐切割小音频 https://pan.baidu.com/s/1hrXxEJU
演讲切割小音频 https://pan.baidu.com/s/1jIrC0F8#list/path=%2F
联系方式
from https://github.com/goodskillprogramer/audiolearning
No comments:
Post a Comment