errors occur when extracting feature
- Dominant language
- Shell
- Stars
- 15.5k
- Forks
- 5.4k
- PR merge metrics
- No merged PRs in 30d
Description
Hi,
I am using egs/sprakbanken_swe/s5 to train model, when comes to command **_steps/make_mfcc.sh --nj 1 --cmd utils/parallel/run.pl data/test exp/make_mfcc/test mfcc_**, the job failed, and the LOG shows:
`Usage: sph2pipe [-h hdr] [-t|-s b:e] [-c 1|2] [-p|-u|-a] [-f typ] infile [outfile]
default conditions (for 'sph2pipe infile'):
* input file contains sphere header
* output full duration of input file
* output all channels from input file
* output same sample coding as input file
* output format is WAV on Wintel machines, SPH elsewhere
* output is written to stdout
optional controls (items bracketed separately above can be combined):
-h hdr -- treat infile as headerless, get sphere info from file 'hdr'
-t b:e -- output portion between b and e sec (floating point)
-s b:e -- output portion between b and e samples (integer)
-c 1 -- only output first channel
-c 2 -- only output second channel
-p -- force conversion to 16-bit linear pcm
-u -- force conversion to 8-bit ulaw
-a -- force conversion to 8-bit alaw
-f typ -- select alternate output header format 'typ'
five types: sph, raw, au, rif(wav), aif(mac)
ERROR (compute-mfcc-feats[5.5.650~4-d9dc2]:Read4ByteTag():wave-reader.cc:56) WaveData: expected 4-byte chunk-name, got read error
[ Stack-Trace: ]
/mnt/e/kaldi/src/lib/libkaldi-base.so(kaldi::MessageLogger::LogMessage() const+0xb42) [0x7f1f2aee6682]
compute-mfcc-feats(kaldi::MessageLogger::LogAndThrow::operator=(kaldi::MessageLogger const&)+0x21) [0x7f1f2bc1520d]
/mnt/e/kaldi/src/lib/libkaldi-feat.so(kaldi::WaveInfo::Read(std::istream&)+0x1c8) [0x7f1f2b5e479e]
/mnt/e/kaldi/src/lib/libkaldi-feat.so(kaldi::WaveData::Read(std::istream&)+0x4e) [0x7f1f2b5e6618]
compute-mfcc-feats(kaldi::WaveHolder::Read(std::istream&)+0x20) [0x7f1f2bc1b846]
compute-mfcc-feats(kaldi::SequentialTableReaderScriptImpl::EnsureObjectLoaded()+0x173) [0x7f1f2bc2324b]
compute-mfcc-feats(kaldi::SequentialTableReaderScriptImpl::Next()+0x27) [0x7f1f2bc23b91]
compute-mfcc-feats(kaldi::SequentialTableReaderScriptImpl::Open(std::__cxx11::basic_string, std::allocator > const&)+0xac) [0x7f1f2bc1862c]
compute-mfcc-feats(kaldi::SequentialTableReader::Open(std::__cxx11::basic_string, std::allocator > const&)+0x1f1) [0x7f1f2bc1bff7]
compute-mfcc-feats(kaldi::SequentialTableReader::SequentialTableReader(std::__cxx11::basic_string, std::allocator > const&)+0x68) [0x7f1f2bc1c352]
compute-mfcc-feats(main+0x845) [0x7f1f2bc1269f]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xe7) [0x7f1f2a331b97]
compute-mfcc-feats(_start+0x2a) [0x7f1f2bc11d7a]
WARNING (compute-mfcc-feats[5.5.650~4-d9dc2]:Read():feat/wave-reader.h:197) Exception caught in WaveHolder::Read(). kaldi::KaldiFatalError
WARNING (compute-mfcc-feats[5.5.650~4-d9dc2]:EnsureObjectLoaded():util/kaldi-table-inl.h:317) Failed to load object from '/mnt/e/kaldi/egs/sprakbanken_swe/s5/../../../tools/sph2pipe_v2.5/sph2pipe -c 1 -p -f wav /mnt/e/kaldi/egs/sprakbanken_swe/s5/data/local/data/download/0468/0468 sv test/Stasjon18/180100/adb_0468/speech/scr0468/18/04681801/r4680017/u0017010.wav |'
WARNING (compute-mfcc-feats[5.5.650~4-d9dc2]:Close():kaldi-io.cc:515) Pipe /mnt/e/kaldi/egs/sprakbanken_swe/s5/../../../tools/sph2pipe_v2.5/sph2pipe -c 1 -p -f wav /mnt/e/kaldi/egs/sprakbanken_swe/s5/data/local/data/download/0468/0468 sv test/Stasjon18/180100/adb_0468/speech/scr0468/18/04681801/r4680017/u0017010.wav | had nonzero return status 256`
the wav file mentioned in the warning message is exist.
Anyone have any idea about this error?
Thank you!
Contributor guide
No contributing guide indexed for this repository
Research direction
Start by rerunning the reported steps/make_mfcc.sh command in egs/sprakbanken_swe/s5 and inspect the referenced WAV path and sph2pipe invocation. Read the compute-mfcc-feats and WaveData error output to determine why the pipe does not produce readable WAV data. Done means the failure is reproducible and has a documented diagnosis or verified fix.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp, shell
- Domain
- audio-video-rtc
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100