how do I use audios larger than 4 seconds? about speech-emotion-analyzer

[MiteshPuthran]

Hey @marvinpx42 , are you talking about the training process or using the trained model on a audio file?

[Rapaix]

thanks for replying, from the trained model.

[MiteshPuthran]

The trained model can't take more than 4 seconds as an input. It takes exactly 216 features as an input due to the sampling rate specified during training. You would have to train your own model using higher sampling rate. Hope this helps.