For example, if I have 3 audio classes including

  1. Ambulance Siren
  2. Police Car Siren
  3. Firetruck Siren

assuming these 3 classes could be distinguished by humans. If I just want the model to classify all these sounds as "Siren" sound only. What approach gives the better performance if I:

  1. Group these classes together into 1 class (Siren sound). and merge all datasets together.
  2. Separate these classes into their own individual categories.

Generally speaking, if the requirement is just to classify those sounds as "siren", you will want to group everything as a "siren" sound.

However, if some other sounds (example: a whale sound) could be confused with a specific siren, the generalisation could be a problem and it would be better to learn every siren sound separately to avoid confusion with other sounds. Then you can group the sirens in one.

As any recongnition problem, we have to take into account similarities in all data in order to tune the neural network correctly.

Hope this answers your question,



Your Answer

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

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