Our task is to feed the audio data into the deep learning tensor flow model in the form of image representation(or graphical,waveform). Question is what is the best way of representing audio in image format??

The representation must focus on words rather than voices.i.e., The similarity of the words must be high than the voices, when two person are speaking the same word.

  • $\begingroup$ There is no way of "...representing audio in image format". The closest you could get to this is the spectrogram but I would not seriously consider its use as an "image". Do you think you could clarify a bit? $\endgroup$ – A_A Feb 13 '20 at 11:48
  • $\begingroup$ @A_A Thank you. Considering audio as a 1d array. We need to represent in a 2d format for the deep learning model to recognize pattern. We tried FFT(spectrogram) but the similarity between the voices is higher than that of the words. Our objective isn't that. Kindly Help. Thank you in advance. $\endgroup$ – Roshan VS Feb 14 '20 at 7:56

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.