ISCA Archive Eurospeech 1999
ISCA Archive Eurospeech 1999

Domain adaptation for robust automatic speech recognition in car environments

Rolf Bippus, Alexander Fischer, Volker Stahl

A major obstacle for the migration of automatic speech recog-nition into every-day life products is environmental robustness. Automatic speech recognition systems work reasonably well un-der clean (laboratory) conditions but degrade seriously under real world conditions (e.g. out-door, car). A lot of research work is devoted to increase the environmental robustness of automatic speech recognition systems. A common method is to use clean (office) data as a starting point and simulate the degraded environ-mental situation by additive artificial (e.g. Gaussian) or recorded noise from the real environment [1]. We study the validity of such additive noise experiments with regard to a real noisy environment. With regard to a previously published work on database adaptation we also examine the possible benefit when using models trained in the simulated environment as a starting point for adap-tation ([2]). We present experimental results on data recorded for task-dependent whole word and phoneme modeling in the car envi-ronment on data from the the MoTiV Car Speech Data Collection (CSDC) [3].


doi: 10.21437/Eurospeech.1999-427

Cite as: Bippus, R., Fischer, A., Stahl, V. (1999) Domain adaptation for robust automatic speech recognition in car environments. Proc. 6th European Conference on Speech Communication and Technology (Eurospeech 1999), 1943-1946, doi: 10.21437/Eurospeech.1999-427

@inproceedings{bippus99_eurospeech,
  author={Rolf Bippus and Alexander Fischer and Volker Stahl},
  title={{Domain adaptation for robust automatic speech recognition in car environments}},
  year=1999,
  booktitle={Proc. 6th European Conference on Speech Communication and Technology (Eurospeech 1999)},
  pages={1943--1946},
  doi={10.21437/Eurospeech.1999-427}
}