BUT System for the Second DIHARD Speech Diarization Challenge
| Autor: | Lukas Burget, Ondrej Novotny, Johan Rohdin, Oldrich Plchot, Hossein Zeinali, Shuai Wang, Federico Landini, Katerina Zmolikova, Pavel Matejka, Ladislav Mosner, Anna Silnova, Mireia Diez |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
Speaker diarisation
Bayes' theorem Channel (digital image) Computer science Audio and Speech Processing (eess.AS) Speech recognition FOS: Electrical engineering electronic engineering information engineering Inference Hidden Markov model Cluster analysis Electrical Engineering and Systems Science - Audio and Speech Processing |
| Zdroj: | Web of Science ICASSP |
| Popis: | This paper describes the winning systems developed by the BUT team for the four tracks of the Second DIHARD Speech Diarization Challenge. For tracks 1 and 2 the systems were mainly based on performing agglomerative hierarchical clustering (AHC) of x-vectors, followed by another x-vector clustering based on Bayes hidden Markov model and variational Bayes inference. We provide a comparison of the improvement given by each step and share the implementation of the core of the system. For tracks 3 and 4 with recordings from the Fifth CHiME Challenge, we explored different approaches for doing multi-channel diarization and our best performance was obtained when applying AHC on the fusion of per channel probabilistic linear discriminant analysis scores. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|