Table 5 False rejection rates for OOV words when using Kaldi-based ASR with N-gram LM
From: Dynamic out-of-vocabulary word registration to language model for speech recognition
Position | Freq. | 2 mora | 3 mora | 4 mora | 5 mora | |
|---|---|---|---|---|---|---|
ONE | – | – | 1 | .977 | .643 | .5 |
ALL | – | – | .313 | .174 | .089 | .022 |
500r | Random | 500 | .783 | .384 | .232 | .109 |
1kr | Random | 1000 | .687 | .314 | .268 | .087 |
2kr | Random | 2000 | .566 | .233 | .268 | .087 |
5kr | Random | 5000 | .537 | .233 | .25 | .087 |
500c | POS | 500 | .747 | .314 | .25 | .065 |
1kc | POS | 1000 | .614 | .291 | .232 | .022 |
2kc | POS | 2000 | .506 | .233 | .214 | .043 |
5kc | POS | 5000 | .390 | .233 | .214 | .065 |