Thank you so much for your feedback ... You are absolutely correct ... Usually, for publishing, people use any dataset, for example in Google they usually use their own huge dataset for the speaker-dependent scenario and not the NIST. But I accept the comparison using NIST dataset could showcase the work in a better way. Unfortunately, I am not working on that project anymore so I have no time for doing that but I am open to collaboration for this effort if anyone is interested to continue.
Thank you so much for your feedback ... We will post very soon about the queue runners. However, since this open source project tries to provide simple codes, in the beginning, our emphasis is one feed_dict simple method although it is suboptimal.
We leveraged 3D convolutional architecture for creating the speaker model in order to simultaneously capturing the speech-related and temporal information from the speakers' utterances.
This open source project is aimed to provide simple and ready-to-use tutorials for TensorFlow. The explanations are present in the wiki associated with this repository. Each tutorial has a source code and its documetation.