ISCA Archive Interspeech 2014
ISCA Archive Interspeech 2014

An initial investigation of long-term adaptation for meeting transcription

X. Chen, Mark J. F. Gales, Kate M. Knill, Catherine Breslin, Langzhou Chen, K. K. Chin, Vincent Wan

Meeting transcription is a very useful and challenging task. The majority of research to date has focused on individual meeting, or only a small group of meetings. In many practical deployments, multiple related meetings will take place over a long period of time. This paper describes an initial investigation of how this long-term data can be used to improve meeting transcription. A corpus of technical meetings, using a single microphone array, was collected over a two year period, yielding a total of 179 hours of meeting data. Baseline systems using deep neural network acoustic models, in both Tandem and Hybrid configurations, and neural network-based language models are described. The impact of supervised and unsupervised adaptation of the acoustic models is then evaluated, as well as the impact of improved language models.

doi: 10.21437/Interspeech.2014-253

Cite as: Chen, X., Gales, M.J.F., Knill, K.M., Breslin, C., Chen, L., Chin, K.K., Wan, V. (2014) An initial investigation of long-term adaptation for meeting transcription. Proc. Interspeech 2014, 954-958, doi: 10.21437/Interspeech.2014-253

  author={X. Chen and Mark J. F. Gales and Kate M. Knill and Catherine Breslin and Langzhou Chen and K. K. Chin and Vincent Wan},
  title={{An initial investigation of long-term adaptation for meeting transcription}},
  booktitle={Proc. Interspeech 2014},