怎样根据声音文件,自动找出声音中每句话的起始时间点和终止时间点? (200分)

比较难!如果有背景声如何?
 
我有一个NCTWavPlayer的OCX控件,不知道大家有没有兴趣
 
to yostgxf
这个控件的功能怎样?
 
功能是很强,可惜是OCX控件,一般使用很不错(录/放/转换/编辑等),但没有具体研究
 
NCTWavPlayer ActiveX Control
Version 1.08


The NCTWavPlayer control is a visual editor of audio files, which allows the user to perform many operations with waveform audio data.

With NCTWavPlayer control you can:

Open and display in special window following types of audio files:
- WAV, in any format, supporting by ACM drivers, installing in your Windows system, including compressing (ADPCM, GSM and others);
- MP3 (if installed Fraunhofer IIS MPEG Layer-3 Codec);
- VOX (Dialogic ADPCM);
- RAW.
Play audio file or any part of it.
Record audio file from microphone or other device.
Edit audio file visually (Cut, Copy, Paste, PasteFromFile, Mix, MixFromFile, Amplify, Invert, Reverse, Normalize, Stretch, Echo).
Convert audio files using ACM drivers.
Change waveform of audio file using GetDataString, PutDataString methods or Value property.
You can insert NCTWavPlayer control in your application, supporting ActiveX controls, such as Visual Basic, Visual C++, Visual FoxPro, Delphi, C++ Builder, and others.
 
各位富豪:
我逐一拜读了上述内容,觉得你们都是多媒体这方面的高手,我有一个问题,不知各位能否给与解答:两个MP3格式的文件应该如何用程序的方法识别其内容的一致性?
 
我想,通过傅立叶变换应该可以吧
 
把声音文件转化成波型然后根据波形来判断,用windows的录音机编辑声音时候,我是这样
取得
 
陈晨:
   你是怎样实现的,能说的详细些吗?最好能有代码!
   怎样联系?laj001@126.com
 
你用傅立叶变换求出人说话时的频率,如果没有说话的时候频率应该是0左右,如果有背景音的话先滤波看看行不行,如果需要傅立叶变换函数的话可以上www.wzlab.com下载,试试效果
 
傅立叶这个词我已经在多处都有所耳闻,要研究一下..
 
满电平的10%
 

Similar threads

S
回复
0
查看
1K
SUNSTONE的Delphi笔记
S
S
回复
0
查看
969
SUNSTONE的Delphi笔记
S
S
回复
0
查看
3K
SUNSTONE的Delphi笔记
S
S
回复
0
查看
2K
SUNSTONE的Delphi笔记
S
D
回复
0
查看
2K
DelphiTeacher的专栏
D
顶部