testadmin
/
Parakeet
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Merge pull request #148 from yt605155624/add_typehint 

add sclite test for text_frontend
This commit is contained in:
Hui Zhang 2021-08-22 22:33:28 -05:00 committed by GitHub
parent b9a30eab9c 36604b4e41
commit c4615e3bba
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
5 changed files with 36 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
examples/text_frontend/README.md
View File

@ -11,6 +11,10 @@ For text normalization, the test data is `data/textnorm_test_cases.txt`, we use
We use `CER` as evaluation criterion.
## Start
If you want to use sclite to get more detail information of WER, you should run the command below to make sclite first.
```bash
./make_sclite.sh
```
Run the command below to get the results of test.
```bash
./run.sh

13
examples/text_frontend/make_sclite.sh Executable file
View File

@ -0,0 +1,13 @@
#!/bin/bash
if [ ! -d "./SCTK" ];then
echo "Clone SCTK ..."
git clone https://github.com/usnistgov/SCTK
echo "Clone SCTK done!"
fi
if [ ! -d "./SCTK/bin" ];then
echo "Start make SCTK ..."
pushd SCTK && make config && make all && make check && make install && make doc && popd
echo "SCTK make done!"
fi

19
examples/text_frontend/run.sh
View File

@ -1,14 +1,25 @@
#!/bin/bash
USE_SCLITE=true
# test g2p
echo "Start get g2p test data."
echo "Start get g2p test data ..."
python3 get_g2p_data.py --root-dir=~/datasets/BZNSYP --output-dir=data/g2p
echo "Start test g2p."
echo "Start test g2p ..."
python3 test_g2p.py --input-dir=data/g2p --output-dir=exp/g2p
# test text normalization
echo "Start get text normalization test data."
echo "Start get text normalization test data ..."
python3 get_textnorm_data.py --test-file=data/textnorm_test_cases.txt --output-dir=data/textnorm
echo "Start test text normalization."
echo "Start test text normalization ..."
python3 test_textnorm.py --input-dir=data/textnorm --output-dir=exp/textnorm
# whether use sclite to get more detail information of WER
if [ "$USE_SCLITE" = true ];then
echo "Start sclite g2p ..."
./SCTK/bin/sclite -i wsj -r ./exp/g2p/text.ref.clean trn -h ./exp/g2p/text.g2p trn -e utf-8 -o all
echo
echo "Start sclite textnorm ..."
./SCTK/bin/sclite -i wsj -r ./exp/textnorm/text.ref.clean trn -h ./exp/textnorm/text.tn trn -e utf-8 -o all
fi

4
examples/text_frontend/test_g2p.py
View File

@ -47,8 +47,8 @@ def get_avg_wer(raw_dict, ref_dict, frontend, output_dir):
gt_phones = [phn for phn in gt_phones if phn not in SILENCE_TOKENS]
gt_phones = " ".join(gt_phones)
g2p_phones = " ".join(g2p_phones)
wf_ref.write(utt_id + " " + gt_phones + "\n")
wf_g2p.write(utt_id + " " + g2p_phones + "\n")
wf_ref.write(gt_phones + "(baker_" + utt_id + ")" + "\n")
wf_g2p.write(g2p_phones + "(baker_" + utt_id + ")" + "\n")
edit_distance, ref_len = word_errors(gt_phones, g2p_phones)
edit_distances.append(edit_distance)
ref_lens.append(ref_len)

4
examples/text_frontend/test_textnorm.py
View File

@ -43,8 +43,8 @@ def get_avg_cer(raw_dict, ref_dict, text_normalizer, output_dir):
gt_text = del_en_add_space(gt_text)
textnorm_text = del_en_add_space(textnorm_text)
wf_ref.write(text_id + " " + gt_text + "\n")
wf_tn.write(text_id + " " + textnorm_text + "\n")
wf_ref.write(gt_text + "(" + text_id + ")" + "\n")
wf_tn.write(textnorm_text + "(" + text_id + ")" + "\n")
edit_distance, ref_len = char_errors(gt_text, textnorm_text)
edit_distances.append(edit_distance)
ref_lens.append(ref_len)