报出调试信息:
Reads in wav file(s) and simulates online decoding.
Writes integerized-text and .ali files for WER computation. Utterance segmentation is done on-the-fly.
Feature splicing/LDA transform is used, if the optional(last) argument is given.
Otherwise delta/delta-delta(i.e. 2-nd order) features are produced.
Caution: the last few frames of the wav file may not be decoded properly.
Hence, don't use one wav file per utterance, but rather use one wav file per show.
主要原因是online_data/run.sh里面的参数配置出错,可以参见:https://github.com/kaldi-asr/kaldi/blob/master/src/onlinebin/online-wav-gmm-decode-faster.cc
由于参数没有正确换行
例如使用tri1 模型,正确参数配置如下:
online-wav-gmm-decode-faster --verbose=1 --rt-min=0.8 --rt-max=0.85\(一定注意这里\前面没有空格)
--max-active=4000 --beam=12.0 --acoustic-scale=0.0769 \
scp:$decode_dir/input.scp $ac_model/final.mdl $ac_model/HCLG.fst \
$ac_model/words.txt '1:2:3:4:5' ark,t:$decode_dir/trans.txt \
ark,t:$decode_dir/ali.txt $trans_matrix;;