Your English writing platform
Discover LudwigExact(3)
In this article, we propose a confusion-driven method to generate phone clusters.
CDPC + PP: the discriminative combination of the proposed confusion-driven phone clusters (CDPC), with phone posteriors (PP).
In this second matrix, we can easily distinguish several sets (clusters) of phones where confusion between all the elements of the cluster is much higher than between other phones or phone clusters.
Similar(57)
One question remains unanswered: how can confusion-driven phone clustering enhance phone recognition?
Figure 5 Hierarchical phone clustering dendrogram using the average distance between clusters ( cophenetic correlation coefficient = 0.873).
The hierarchical phone clustering dendrogram in Figure 5 gives rise to the BPC9.
a"Confusion-driven" is used since the phone clustering is a function of a confusion matrix.
Model-driven methods and confusion-drivena methods are then the two major categories of data-driven phone clustering algorithms.
Furthermore, the scope of the current proposal extended to applying the confusion-driven phone clustering method to a phone recognition system.
The results, using TIMIT data, show that the proposed confusion-driven phone clustering method is an attractive alternative to the approaches based on human knowledge.
This article has described a phone clustering method based on a phone similarity metric which is computed from a confusion matrix generated from the output of a phone recognition system.
Write better and faster with AI suggestions while staying true to your unique style.
Since I tried Ludwig back in 2017, I have been constantly using it in both editing and translation. Ever since, I suggest it to my translators at ProSciEditing.

Justyna Jupowicz-Kozak
CEO of Professional Science Editing for Scientists @ prosciediting.com