Sorry, you need to enable JavaScript to visit this website.

SP-P5: Speaker Diarization & Identification

Optimizing Bayesian HMM Based x-vector Clustering for theSecond DIHARD Speech Diarization Challenge

Paper Details

Authors:
Mireia Diez, Lukas Burget, Federico Landini, Shuai Wang, Honza Cernocky
Submitted On:
21 May 2020 - 9:13am
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

ICASSP2020_DIHARD_BHMM_Slides.pdf

(15)

Subscribe

[1] Mireia Diez, Lukas Burget, Federico Landini, Shuai Wang, Honza Cernocky, "Optimizing Bayesian HMM Based x-vector Clustering for theSecond DIHARD Speech Diarization Challenge", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5427. Accessed: Jul. 07, 2020.
@article{5427-20,
url = {http://sigport.org/5427},
author = {Mireia Diez; Lukas Burget; Federico Landini; Shuai Wang; Honza Cernocky },
publisher = {IEEE SigPort},
title = {Optimizing Bayesian HMM Based x-vector Clustering for theSecond DIHARD Speech Diarization Challenge},
year = {2020} }
TY - EJOUR
T1 - Optimizing Bayesian HMM Based x-vector Clustering for theSecond DIHARD Speech Diarization Challenge
AU - Mireia Diez; Lukas Burget; Federico Landini; Shuai Wang; Honza Cernocky
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5427
ER -
Mireia Diez, Lukas Burget, Federico Landini, Shuai Wang, Honza Cernocky. (2020). Optimizing Bayesian HMM Based x-vector Clustering for theSecond DIHARD Speech Diarization Challenge. IEEE SigPort. http://sigport.org/5427
Mireia Diez, Lukas Burget, Federico Landini, Shuai Wang, Honza Cernocky, 2020. Optimizing Bayesian HMM Based x-vector Clustering for theSecond DIHARD Speech Diarization Challenge. Available at: http://sigport.org/5427.
Mireia Diez, Lukas Burget, Federico Landini, Shuai Wang, Honza Cernocky. (2020). "Optimizing Bayesian HMM Based x-vector Clustering for theSecond DIHARD Speech Diarization Challenge." Web.
1. Mireia Diez, Lukas Burget, Federico Landini, Shuai Wang, Honza Cernocky. Optimizing Bayesian HMM Based x-vector Clustering for theSecond DIHARD Speech Diarization Challenge [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5427

Speaker Diarization with Session-level Speaker Embedding Refinement using Graph Neural Networks


Deep speaker embedding models have been commonly used as a building block for speaker diarization systems; however, the speaker embedding model is usually trained according to a global loss defined on the training data, which could be sub-optimal for distinguishing speakers locally in a specific meeting session. In this work we present the first use of graph neural networks (GNNs) for the speaker diarization problem, utilizing a GNN to refine speaker embeddings locally using the structural information between speech segments inside each session.

Paper Details

Authors:
Jixuan Wang, Xiong Xiao, Jian Wu, Ranjani Ramamurthy, Frank Rudzicz, Michael Brudno
Submitted On:
13 May 2020 - 8:20pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Slides for the paper "Speaker Diarization with Session-level Speaker Embedding Refinement using Graph Neural Networks"

(20)

Subscribe

[1] Jixuan Wang, Xiong Xiao, Jian Wu, Ranjani Ramamurthy, Frank Rudzicz, Michael Brudno, "Speaker Diarization with Session-level Speaker Embedding Refinement using Graph Neural Networks", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5168. Accessed: Jul. 07, 2020.
@article{5168-20,
url = {http://sigport.org/5168},
author = {Jixuan Wang; Xiong Xiao; Jian Wu; Ranjani Ramamurthy; Frank Rudzicz; Michael Brudno },
publisher = {IEEE SigPort},
title = {Speaker Diarization with Session-level Speaker Embedding Refinement using Graph Neural Networks},
year = {2020} }
TY - EJOUR
T1 - Speaker Diarization with Session-level Speaker Embedding Refinement using Graph Neural Networks
AU - Jixuan Wang; Xiong Xiao; Jian Wu; Ranjani Ramamurthy; Frank Rudzicz; Michael Brudno
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5168
ER -
Jixuan Wang, Xiong Xiao, Jian Wu, Ranjani Ramamurthy, Frank Rudzicz, Michael Brudno. (2020). Speaker Diarization with Session-level Speaker Embedding Refinement using Graph Neural Networks. IEEE SigPort. http://sigport.org/5168
Jixuan Wang, Xiong Xiao, Jian Wu, Ranjani Ramamurthy, Frank Rudzicz, Michael Brudno, 2020. Speaker Diarization with Session-level Speaker Embedding Refinement using Graph Neural Networks. Available at: http://sigport.org/5168.
Jixuan Wang, Xiong Xiao, Jian Wu, Ranjani Ramamurthy, Frank Rudzicz, Michael Brudno. (2020). "Speaker Diarization with Session-level Speaker Embedding Refinement using Graph Neural Networks." Web.
1. Jixuan Wang, Xiong Xiao, Jian Wu, Ranjani Ramamurthy, Frank Rudzicz, Michael Brudno. Speaker Diarization with Session-level Speaker Embedding Refinement using Graph Neural Networks [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5168

MULTISTREAM DIARIZATION FUSION USING THE MINIMUM VARIANCE BAYESIAN INFORMATION CRITERION

Paper Details

Authors:
Submitted On:
12 April 2018 - 3:24pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Poster_ICASSP_2018.pdf

(153)

Subscribe

[1] , "MULTISTREAM DIARIZATION FUSION USING THE MINIMUM VARIANCE BAYESIAN INFORMATION CRITERION", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2476. Accessed: Jul. 07, 2020.
@article{2476-18,
url = {http://sigport.org/2476},
author = { },
publisher = {IEEE SigPort},
title = {MULTISTREAM DIARIZATION FUSION USING THE MINIMUM VARIANCE BAYESIAN INFORMATION CRITERION},
year = {2018} }
TY - EJOUR
T1 - MULTISTREAM DIARIZATION FUSION USING THE MINIMUM VARIANCE BAYESIAN INFORMATION CRITERION
AU -
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2476
ER -
. (2018). MULTISTREAM DIARIZATION FUSION USING THE MINIMUM VARIANCE BAYESIAN INFORMATION CRITERION. IEEE SigPort. http://sigport.org/2476
, 2018. MULTISTREAM DIARIZATION FUSION USING THE MINIMUM VARIANCE BAYESIAN INFORMATION CRITERION. Available at: http://sigport.org/2476.
. (2018). "MULTISTREAM DIARIZATION FUSION USING THE MINIMUM VARIANCE BAYESIAN INFORMATION CRITERION." Web.
1. . MULTISTREAM DIARIZATION FUSION USING THE MINIMUM VARIANCE BAYESIAN INFORMATION CRITERION [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2476