Speaker Detection in the Wild: Lessons Learned from JSALT 2019

Leibny Paola Garcia Perera, Jesus Villalba, Herve Bredin, Jun Du, Diego Castan, Alejandrina Cristia, Latane Bullock, Ling Guo, Koji Okabe, Phani Sankar Nidadavolu, Saurabh Kataria, Sizhu Chen, Leo Galmant, Marvin Lavechin, Lei Sun, Marie-Philippe Gill, Bar Ben-Yair, Sajjad Abdoli, Xin Wang, Wassim Bouaziz, Hadrien Titeux, Emmanuel Dupoux, Kong Aik Lee, Najim Dehak


This paper presents the problems and solutions addressed at theJSALT workshop when using a single microphone for speaker detection in adverse scenarios. The main focus was to tackle a wide range of conditions that go from meetings to wild speech. We describe the research threads we explored and a set of modules that was successful for these scenarios. The ultimate goal was to explore speaker detection; but our first finding was that an effective diarization improves detection, and not having a diarization stage impoverishes the performance. All the different configurations of our research agree on this fact and follow a main backbone that includes diarization asa previous stage. With this backbone, we analyzed the following problems: voice activity detection, how to deal with noisy signals, domain mismatch, how to improve the clustering; and the overall impact of previous stages in the final speaker detection. In this pa-per, we show partial results for speaker diarizarion to have a better understanding of the problem and we present the final results for speaker detection


 DOI: 10.21437/Odyssey.2020-59

Cite as: Garcia Perera, L.P., Villalba, J., Bredin, H., Du, J., Castan, D., Cristia, A., Bullock, L., Guo, L., Okabe, K., Nidadavolu, P.S., Kataria, S., Chen, S., Galmant, L., Lavechin, M., Sun, L., Gill, M., Ben-Yair, B., Abdoli, S., Wang, X., Bouaziz, W., Titeux, H., Dupoux, E., Lee, K.A., Dehak, N. (2020) Speaker Detection in the Wild: Lessons Learned from JSALT 2019. Proc. Odyssey 2020 The Speaker and Language Recognition Workshop, 415-422, DOI: 10.21437/Odyssey.2020-59.


@inproceedings{Garcia Perera2020,
  author={Leibny Paola  {Garcia Perera} and Jesus Villalba and Herve Bredin and Jun Du and Diego Castan and Alejandrina Cristia and Latane Bullock and Ling Guo and Koji Okabe and Phani Sankar Nidadavolu and Saurabh Kataria and Sizhu Chen and Leo Galmant and Marvin Lavechin and Lei Sun and Marie-Philippe Gill and Bar Ben-Yair and Sajjad Abdoli and Xin Wang and Wassim Bouaziz and Hadrien Titeux and Emmanuel Dupoux and Kong Aik Lee and Najim Dehak},
  title={{Speaker Detection in the Wild: Lessons Learned from JSALT 2019}},
  year=2020,
  booktitle={Proc. Odyssey 2020 The Speaker and Language Recognition Workshop},
  pages={415--422},
  doi={10.21437/Odyssey.2020-59},
  url={http://dx.doi.org/10.21437/Odyssey.2020-59}
}