Files
FunASR/runtime/tools/train_compile_ngram.sh
T
freedakgmail 6116b1f3c6
Update API Documentation / build-api-docs (push) Has been cancelled
Initial commit: FunASR Speech Recognition Toolkit
Add complete FunASR codebase including models, runtime, and documentation.
2026-07-09 22:38:58 +08:00

22 lines
649 B
Bash

chmod +x fst/*
[ -f path.sh ] && . ./path.sh
# download train corpus and lexicon
wget https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/requirements/lm.tar.gz
tar -zxvf lm.tar.gz
# train lm, make sure that srilm is installed
bash fst/train_lms.sh
# generate lexicon
python3 fst/generate_lexicon.py lm/corpus.dict lm/lexicon.txt lm/lexicon.out
# Compile the lexicon and token FSTs
fst/compile_dict_token.sh lm lm/tmp lm/lang
# Compile the language-model FST and the final decoding graph TLG.fst
fst/make_decode_graph.sh lm lm/lang || exit 1;
# Collect resource files required for decoding
fst/collect_resource_file.sh lm lm/resource