I will do voice activity detection and speech enhancement based deep neural network. However, I don't know whether to do this via matlab or pyhton. In which programming language can I find more ready-made code on this subject? Which one do you recommend?

  • 2
    $\begingroup$ of course python $\endgroup$
    – ZR Han
    Commented Oct 13, 2021 at 9:33
  • $\begingroup$ this is purely a question for opinions, and those are explicitly off-topic here. $\endgroup$ Commented Oct 13, 2021 at 11:14
  • $\begingroup$ of course COBOL. $\endgroup$ Commented Oct 14, 2021 at 6:47

1 Answer 1


The language choice depends on many factors.
For instance, are you after developing low level features of DNN or using existing building blocks?

Most advanced and popular Deep Neural Networks (DNN) Frameworks are nativly integrated into Python though they are mostly implemented using different low level language (C++ mainly). Those include PyTorch and TensorFlow.

They have interfaces to other languages as well and there are other good competitors (I like KNet.jl which is based on Julia).

The Python DNN universe is far more rich than MATLAB's. So the between choosing between the 2 the choice is pretty easy unless there is some other factors such as:

  1. Code Generation
    MATLAB's implementation supports native code generation which can be integrated into production C / C++ code (Or any code with interface to C).
  2. Existing Code in MATLAB
    In case you have an existing code which is implemented with MATLAB than the easier integration is a big factor.

MathWorks has improved its DNN toolbox greatly and it has most common features but still it is not as flexible and widely used as those in Python (Also not free).

  • 1
    $\begingroup$ The Knet.jl package looks like PyTorch for Julia. $\endgroup$
    – David
    Commented Nov 21, 2021 at 17:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.