Can anyone recommend a decent DSP/speech library i

问题:

Google returns too much results, although SPUC caught my attention. Is there a standard recommended library like OpenCV for vision? The necessary features would be:

• Free Open Source
• filter design (Butterworth, Chebyshev, etc)
• FFT
• if possible, some speech processing features, like MFCC computation, although that's secondary, as I could use SPTK (sp-tk.sourceforge.net) for that part.

回答1:

The Synthesis Toolkit, https://ccrma.stanford.edu/software/stk/, has a class that can model different Phonemes. It also has tools for all sorts of DSP including different types of filters. I recommend checking it out as it will be a fantastic learning experience no matter what you use it for.

回答2:

I've used CMU's Sphinx library with success.