Table 4.

CERs (%) for each method when using RNN-based ASR models with JNAS as the target domain, where [X] indicates an ASR model trained with dataset X and Y+Z indicates a single ASR model trained with datasets Y and Z. [Y]+[Z], on the other hand, represents the integration of the outputs of ASR models trained with datasets Y and Z, respectively. The ROVER method is theoretically applicable to two models, but because majority voting fails and results in a tie, this study reports results only for the case of integration of three models

ASR modelNIntegration method
w/o LMSFDRAROVERProposed
[APS]126.323.922.4
[SPS]122.021.719.9
[CV]131.330.529.9
[APS+SPS]118.916.615.9
[APS+SPS+CV]116.014.513.7
[APS]+[SPS]219.417.916.6
[APS]+[SPS]+[CV]319.618.425.516.4

or Create an Account

Close Modal
Close Modal