Thank you so much for checking our project!! Hope you liked it.
Regarding your concerns, we will be adding a copyright checker feature to ensure originality. We will be training our model further. It will have popular singer’s lyrics & songs as dataset! We will utilise existing word embedding models like word2vec’s cosine difference(one of the ways) to find similarity between words in ml generated lyrics and lyrics in data set!
If you liked our project, you can vote for us here !
Regards