Sixth European Conference on Speech Communication and Technology
(EUROSPEECH'99)

Budapest, Hungary
September 5-9, 1999

Finding Consensus Among Words: Lattice-Based Word Error Minimization

Lidia Mangu (1), Eric Brill (1), Andreas Stolcke (2)

(1) Department of Computer Science, Johns Hopkins University, Baltimore, MD, USA
(2) Speech Technology and Research Laboratory, SRI International, Menlo Park, CA, USA

We describe a new algorithm for finding the hypothesis in a recognition lattice that is expected to minimize the word error rate (WER). Our approach thus overcomes the mismatch between the word-based performance metric and the standard MAP scoring paradigm that is sentence-based, and that can lead to sub-optimal recognition results. To this end we first find a complete alignment of all words in the recognition lattice, identifying mutually supporting and competing word hypotheses. Finally, a new sentence hypothesis is formed by concatenating the words with maximal posterior probabilities. Experimentally, this approach leads to a significant WER reduction in a large vocabulary recognition task.


Full Paper (PDF)   Gnu-Zipped Postscript

Bibliographic reference.  Mangu, Lidia / Brill, Eric / Stolcke, Andreas (1999): "Finding consensus among words: lattice-based word error minimization", In EUROSPEECH'99, 495-498.