mirror of
https://github.com/k2-fsa/icefall.git
synced 2025-09-19 05:54:20 +00:00
Update docs/source/decoding-with-langugage-models/LODR.rst
Co-authored-by: Fangjun Kuang <csukuangfj@gmail.com>
This commit is contained in:
parent
5ff647e226
commit
78fec8ef6f
@ -47,7 +47,7 @@ shallow fusion is the subtraction of the source domain LM.
|
||||
Some works treat the predictor and the joiner of the neural transducer as its internal LM. However, the LM is
|
||||
considered to be weak and can only capture low-level language information. Therefore, `LODR <https://arxiv.org/abs/2203.16776>`__ proposed to use
|
||||
a low-order n-gram LM as an approximation of the ILM of the neural transducer. This leads to the following formula
|
||||
during decoding for RNNT model:
|
||||
during decoding for transducer model:
|
||||
|
||||
.. math::
|
||||
|
||||
|
Loading…
x
Reference in New Issue
Block a user