mirrors/icefall
1
0
Fork 0
mirror of https://github.com/k2-fsa/icefall.git synced 2025-08-14 12:32:20 +00:00
Code Issues Packages Projects Releases Wiki Activity

fixes

Browse Source
This commit is contained in:
Piotr Żelasko 2022-01-21 19:20:57 +00:00
parent e76de3ba59
commit cb329d1342
2 changed files with 8 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
egs/fisher_swbd/ASR/local/prepare_lang_g2pen.py
View File

@ -17,21 +17,10 @@
"""
This script takes as input a lexicon file "data/lang_phone/lexicon.txt"
consisting of words and tokens (i.e., phones) and does the following:
1. Add disambiguation symbols to the lexicon and generate lexicon_disambig.txt
2. Generate tokens.txt, the token table mapping a token to a unique integer.
3. Generate words.txt, the word table mapping a word to a unique integer.
4. Generate L.pt, in k2 format. It can be loaded by
d = torch.load("L.pt")
lexicon = k2.Fsa.from_dict(d)
5. Generate L_disambig.pt, in k2 format.
This script takes as input a wors.txt file "data/lang_phone/words.txt"
consisting of words and their IDs and creates a lexicon with g2p_en python package
(it's CMUdict based). It also creates rest of the files typically expected in a lang
dir, including L.pt and Linv.pt.
"""
import argparse
import math

8
egs/fisher_swbd/ASR/prepare.sh
View File

@ -103,10 +103,10 @@ if [ $stage -le 4 ] && [ $stop_stage -ge 4 ]; then
# Combine Fisher and SWBD recordings and supervisions
if $swbd_only; then
cp data/manifests/swbd/swbd_recordings.jsonl \
data/manifests/fisher-swbd_recordings.jsonl.gz
cp data/manifests/swbd/swbd_supervisions.jsonl \
data/manifests/fisher-swbd_supervisions.jsonl.gz
gunzip -c data/manifests/swbd/swbd_recordings.jsonl \
> data/manifests/fisher-swbd_recordings.jsonl.gz
gunzip -c data/manifests/swbd/swbd_supervisions.jsonl \
> data/manifests/fisher-swbd_supervisions.jsonl.gz
else
lhotse combine \
data/manifests/fisher/recordings.jsonl.gz \