This paper describes the winning systems developed by the BUT team for the four tracks of the Second DIHARD Speech Diarization Challenge, with source code available

We proposed the text-adptation speaker verification task and an intital solution called Speaker-text factorization network, which could deal with different text-mismatch conditions

This paper describes the winning systems developed by the BUT team for the two tracks of the First VoxSRC Speaker Recognition Challenge

We proposed the segment-level representation for phonetic information and the corresponding segment-level multi-task/adversarial training framework, we revisited the usage the phonetic information for the text-independent embedding learning and designed experiments to verify the assumption: For TI-SV, it could be benificial to remove the phonetic variation in the final speaker embeddings

The first attempt to systematically analyze the information encoded in speaker embeddings (prior to x-vector), detailed analysis on x-vectors could be refered to the paper Probing the Information Encoded in X-vectors from JHU