search

tyiannak / pyAudioAnalysis

Python Audio Analysis Library: Feature Extraction, Classification, Segmentation and Applications
Apache License 2.0
5.81k stars 1.19k forks source link

How to get short audio content similarity when I extract the MFCC features? #167

Open Pen0201 opened 6 years ago

Pen0201 commented 6 years ago

If there any function in this package support to compute the similarity of two audio file after MFCC- features extraction. Thanks a lot.

jhulianabarrios commented 6 years ago

??

El mié., 16 may. 2018 19:05, PHE notifications@github.com escribió:

If there any function in this package support to compute the similarity of two audio file after MFCC- features extraction. Thanks a lot.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/tyiannak/pyAudioAnalysis/issues/167, or mute the thread https://github.com/notifications/unsubscribe-auth/Ak8q2ABtWp0Hb-WD7oFUzIP_vTKxprNTks5tzL63gaJpZM4UCN5d .

Pen0201 commented 6 years ago

Sorry for the question is not clear ! I have two different speech audio file(ex.a.wav and b.wav),but they have the same speech contents (ex.hello or how are you ,etc) if there any function in this package to compute the similarity score for this two audio file ? my first step is to extract MFCCs feature (Is it right?) ,and I can not find any function to calculate the similarity score from MFCCs features for next step.

pvanlaar commented 6 years ago

So you want to compare the similarity between the speech found in both files?

If you have the MFCC feature vectors for both, you can take the Euclidean distance. That will be some measure for the similarity. Correct me if I don't understand your approach.

This should be labelled as a question by the way.