KaldiFatalError during decoding phase

[tanzyy96]

I wanted to try out the PASE Kaldi ASR experiment on the TIMIT dataset. The initial training phase was successful. However, during the decoding phase, the code abruptly ended with the following message:

run.pl: job failed, log is in /Project0550/zhiyang/pase/out_folder/dec/scoring/log/best_path.1.1.log

Opening the file:

# lattice-align-phones /Project0550/zhiyang/pase/out_folder/dec/../final.mdl "ark:gunzip -c /Project0550/zhiyang/pase/out_folder/dec/lat.1.gz|" ark:- | lattice-to-ctm-conf --acoustic-scale= --lm-scale=1.0 ark:- /Project0550/zhiyang/pase/out_folder/dec/scoring/1.1.ctm 
# Started at Mon Oct  5 01:28:00 +08 2020
#
lattice-align-phones /Project0550/zhiyang/pase/out_folder/dec/../final.mdl 'ark:gunzip -c /Project0550/zhiyang/pase/out_folder/dec/lat.1.gz|' ark:- 
ERROR (lattice-to-ctm-conf[5.5.809~2-484f57]:ToFloat():parse-options.cc:605) Invalid floating-point option ""

[ Stack-Trace: ]
/Project0550/zhiyang/kaldi/src/lib/libkaldi-base.so(kaldi::MessageLogger::LogMessage() const+0x82c) [0x7fef2a8a62ca]
lattice-to-ctm-conf(kaldi::MessageLogger::LogAndThrow::operator=(kaldi::MessageLogger const&)+0x21) [0x419dc7]
/Project0550/zhiyang/kaldi/src/lib/libkaldi-util.so(kaldi::ParseOptions::ToFloat(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)+0x97) [0x7fef2aad0079]
/Project0550/zhiyang/kaldi/src/lib/libkaldi-util.so(kaldi::ParseOptions::SetOption(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)+0x3fc) [0x7fef2aad3db0]
/Project0550/zhiyang/kaldi/src/lib/libkaldi-util.so(kaldi::ParseOptions::Read(int, char const* const*)+0x37e) [0x7fef2aad4a66]
lattice-to-ctm-conf(main+0x37f) [0x4176b5]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf0) [0x7fef299f9830]
lattice-to-ctm-conf(_start+0x29) [0x417269]

kaldi::KaldiFatalError# Accounting: time=1 threads=1
# Ended (code 255) at Mon Oct  5 01:28:01 +08 2020, elapsed time 1 seconds

Any ideas?