mirrors/icefall
1
0
Fork 0
mirror of https://github.com/k2-fsa/icefall.git synced 2025-12-09 14:05:33 +00:00
Code Issues Packages Projects Releases Wiki Activity

Minor fixes.

Browse Source
This commit is contained in:
Fangjun Kuang 2021-07-20 19:54:12 +08:00
parent d5e0408698
commit 8a72901f3a
1 changed files with 3 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
egs/librispeech/ASR/local/prepare_lang.py
View File

@ -41,9 +41,6 @@ def read_lexicon(filename: str) -> Lexicon:
That is, the first field is a word and the remaining That is, the first field is a word and the remaining
fields are phones. Fields are separated by space(s). fields are phones. Fields are separated by space(s).
We assume that the input lexicon does not contain words:
<eps>, <s>, </s>, !SIL, <SPOKEN_NOISE>, <UNK>.
Args: Args:
filename: filename:
Path to the lexicon.txt Path to the lexicon.txt
@ -51,10 +48,6 @@ def read_lexicon(filename: str) -> Lexicon:
Returns: Returns:
A list of tuples., e.g., [('w', ['p1', 'p2']), ('w1', ['p3, 'p4'])] A list of tuples., e.g., [('w', ['p1', 'p2']), ('w1', ['p3, 'p4'])]
""" """
# ans = ["!SIL", ["SIL"]]
# ans.append(["<SPOKEN_NOISE>", ["SPN"]])
# ans.append(["<UNK>", ["SPN"]])
ans = [] ans = []
with open(filename, "r", encoding="latin-1") as f: with open(filename, "r", encoding="latin-1") as f:
@ -359,7 +352,7 @@ def main():
# whose ilabel is phone2id['#0'] and olable is word2id['#0'] # whose ilabel is phone2id['#0'] and olable is word2id['#0']
# Need to implement it in k2 # Need to implement it in k2
if False: if True:
# Just for debugging, will remove it # Just for debugging, will remove it
torch.save(L.as_dict(), out_dir / "L.pt") torch.save(L.as_dict(), out_dir / "L.pt")
torch.save(L_disambig.as_dict(), out_dir / "L_disambig.pt") torch.save(L_disambig.as_dict(), out_dir / "L_disambig.pt")
@ -368,8 +361,8 @@ def main():
L.aux_labels_sym = k2.SymbolTable.from_file(out_dir / "words.txt") L.aux_labels_sym = k2.SymbolTable.from_file(out_dir / "words.txt")
L_disambig.labels_sym = L.labels_sym L_disambig.labels_sym = L.labels_sym
L_disambig.aux_labels_sym = L.aux_labels_sym L_disambig.aux_labels_sym = L.aux_labels_sym
L.draw(out_dir / "L.svg", title="L") L.draw(out_dir / "L.png", title="L")
L_disambig.draw(out_dir / "L_disambig.svg", title="L_disambig") L_disambig.draw(out_dir / "L_disambig.png", title="L_disambig")
if __name__ == "__main__": if __name__ == "__main__":