# 把所有字符中的大写字母转换成小写字母
def lower(sent):
return sent.lower()
# Read all EN sents
input = 'data/phoenix2014T'
for subset in ['train','dev','test']:
for lang in ['de','gloss']:
with open(input+'/{}.{}'.format(subset,lang), 'r', encoding='utf-8') as f1, \
open(input + '/{}.lower.{}'.format(subset, lang), 'w', encoding='utf-8') as f2:
en_sents = f1.readlines()
for en_sent in en_sents:
lower_en_sent = lower(en_sent)
f2.write(lower_en_sent)