There is a need to classify an audio stream in real-time - using one from thousands of the defined classes. Up to 5000 classes potentially.

What's the best machine learning algorithm for this task?

DTW? HMM? SVM? Some specific DSP algorithms?

  • $\begingroup$ Anything that doesn't require extracting expensive features; something that works with FFTs and MFCCs, such as a modern neural network should be fine. The training complexity is not an issue in your case. $\endgroup$
    – Emre
    Jul 1 '16 at 20:19
  • $\begingroup$ Could give some example of the classes? Is your problem similar to speech recognition? $\endgroup$
    – William
    Jul 1 '16 at 20:24
  • $\begingroup$ Yes, similar to speech recognition, but discovering such things as "car engine", "scream", "human steps", "dog barks" etc. $\endgroup$ Jul 1 '16 at 20:32
  • $\begingroup$ SVM should work too. $\endgroup$
    – n1tk
    Jul 3 '16 at 4:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.