This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: 1: 7 -&gt; e 2: 6 -&gt; a For substitution, o substitution happens much more often than the others. This might be caused by ambiguity of the model for o, which in this case recognizing as incorrect word. The most occurrences of substitution: 1: 20 -&gt; o 2: 7 -&gt; a 3: 6 -&gt; e 4: 6 -&gt; eight The most often falsely recognized: 1: 14 -&gt; oh 2: 12 -&gt; t 3: 10 -&gt; eighty Summary: The result of the test shows that most of the error caused by the incorrect recognition of the model o. This might be caused by insufficient o data in the training set or the nature of o which is similar to other model. To reduce the error rate, we can train with a better or more train data containing o....
View Full Document
This note was uploaded on 02/11/2012 for the course ECE 5526 taught by Professor Staff during the Summer '09 term at FIT.
- Summer '09