C# 或 C++ 中的音频处理
我想创建一个使用 AI 技术的应用程序，允许用户录制歌曲的一部分，然后尝试在 wav 文件数据库中查找该歌曲。
我希望使用一些已经存在的库来进行音频处理部分。那么，您能否推荐 C# 中的任何库，它可以读取 wav 文件、从麦克风获取输入、具有一些音频滤波器（低通、高通、FFT 等），并且还可以绘制音频信号。
我更喜欢用 C# 开发，但如果没有好的音频处理库，我想我也可以用 C++ 工作。据我所知，Mathlab已经具有上述功能，但我无法在我的应用程序中使用它。
I would like to create an application that uses AI techniques and allows the user to record a part of a song and then tries to find that song in a database of wav files.
I would have liked to use some already existing libraries for the audio processing part. So, could you recommend any libraries in C# which can read a wav file, get input from microphone, have some audio filters (low pass, high pass, FFT etc) and maybe have the ability to plot the audio signal as well.
I would prefer to develop in C#, but if there aren't good libraries for audio processing, I guess I could work in C++ as well. As far as I know, Mathlab already has the above mentioned functionalities, but I can't use it in my application.
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（3）
如果您只想读取和记录音频信号，我推荐这个库
NAudio：https://naudio.codeplex.com
但如果你需要信号处理支持，我认为这个更好
CSCORE：https://github.com/filoe/cscore
If you just want to read and record audio signal, I recommend this lib
NAudio: https://naudio.codeplex.com
But if you need a signal processing support, I think this one is much better
CSCORE: https://github.com/filoe/cscore
例如，看看 FMOD ，但不确定它是否可以完成您需要的一切。
Take a look at FMOD for example, not sure if it can do everything you need though.
我不知道有任何公开可用的“音频指纹/匹配”库可以处理短片段。有一个 libfooid 项目，但我认为它需要您录制整首歌曲。此外，音频设备交互不是该库的一部分。您必须使用另一个库来录制音频数据。
I'm not aware of any publicly available "audio fingerprint / matching" libraries that can deal with short snippets. There's a libfooid project but I think it requires you to record the whole song. Also, the audio device interaction is not part of that library. You have to use another library for recording audio data.