Utterance-level Aggregation For Speaker Recognition In The Wild - View it on GitHub