DeepSpeech/data/lm
2019-11-15 13:28:45 +01:00
..
generate_lm.py Switch to --prune 0 0 1 model and move generation code to a script 2019-11-15 13:28:45 +01:00
lm.binary Switch to --prune 0 0 1 model and move generation code to a script 2019-11-15 13:28:45 +01:00
README.rst Switch to --prune 0 0 1 model and move generation code to a script 2019-11-15 13:28:45 +01:00
trie Filter LM by removing very rare words 2019-11-13 17:38:40 +01:00

lm.binary was generated from the LibriSpeech normalized LM training text, available `here <http://www.openslr.org/11>`_\ , using the `generate_lm.py` script (will generate lm.binary in the folder it is run from). KenLM's built binaries must be in your PATH (lmplz, build_binary, filter).

The trie was then generated from the vocabulary of the language model:

.. code-block:: bash

   ./generate_trie ../data/alphabet.txt lm.binary trie