Collaborative Real-Time Speaker Identification for Wearable Systems. Rossi, M., Amft, O., Kusserow, M., & Tröster, G. In PerCom 2010: Proceedings of the 8th Annual IEEE International Conference on Pervasive Computing and Communications, pages 180--189, 2010. IEEE. Acceptance rate: 12%
doi  abstract   bibtex   
We present an unsupervised speaker identification system for personal annotations of conversations and meetings. The system dynamically learns new speakers and recognizes already known speakers using one audio channel and speech-independent modeling. Multiple personal systems could collaborate in robust unsupervised speaker identification and online learning. The system was optimized for real-time operation on a DSP system that can be worn during daily activities. The system was evaluated on the freely available 24-speaker Augmented Multiparty Interaction dataset. For 5s recognition time, the system achieves 81% recognition rate. Collaboration between four identification systems resulted in a performance increase of up to 17%, however even two collaborating systems yield an performance improvement. A prototypical wearable DSP implementation could continuously operate for more than 8hours from a 4.1Ah battery.
@InProceedings{Rossi2010-P_PerCom,
  Title                    = {Collaborative Real-Time Speaker Identification for Wearable Systems.},
  Author                   = {Mirco Rossi and Oliver Amft and Martin Kusserow and Gerhard Tr\"oster},
  Booktitle                = {PerCom 2010: Proceedings of the 8th Annual IEEE International Conference on Pervasive Computing and Communications},
  Year                     = {2010},
  Note                     = {Acceptance rate: 12\%},
  Pages                    = {180--189},
  Publisher                = {IEEE},

  Abstract                 = {We present an unsupervised speaker identification system for personal annotations of conversations and meetings. The system dynamically learns new speakers and recognizes already known speakers using one audio channel and speech-independent modeling. Multiple personal systems could collaborate in robust unsupervised speaker identification and online learning. The system was optimized for real-time operation on a DSP system that can be worn during daily activities. The system was evaluated on the freely available 24-speaker Augmented Multiparty Interaction dataset. For 5s recognition time, the system achieves 81\% recognition rate. Collaboration between four identification systems resulted in a performance increase of up to 17\%, however even two collaborating systems yield an performance improvement. A prototypical wearable DSP implementation could continuously operate for more than 8hours from a 4.1Ah battery.},
  Doi                      = {10.1109/PERCOM.2010.5466976},
  File                     = {Rossi2010-P_PerCom:Rossi2010-P_PerCom.pdf:PDF},
  Owner                    = {oam},
  Timestamp                = {2010/01/14}
}

Downloads: 0