Regarding the project flow

[SivaKanishka]

Hi @woshipzs,

Can you explain about how you structured this project, and any datasets you used for training the model.

Thanks in advance

[woshipzs]

Hi, @SivaKanishka , for this project, the dataset that we used is from a private company. They recorded phone calls between customers and representatives. We parsed those phone call records into a format that we can use in our study. However, the dataset is private and we cannot make it publicly available, The structure of this project is basically two parts: one is to preprocess raw audio data(clean, diarization, separation, *transform into other formats if needed, etc.), the other part is to use these pre-processed data and feed them into cnns then compare the results. Let me know if you have more questions.