alphacep / vosk-api Goto Github PK
View Code? Open in Web Editor NEWOffline speech recognition API for Android, iOS, Raspberry Pi and servers with Python, Java, C# and Node
License: Apache License 2.0
Offline speech recognition API for Android, iOS, Raspberry Pi and servers with Python, Java, C# and Node
License: Apache License 2.0
Hi, i found there is no script to add lm-rescoring. Can you please guide me on how to perform rescoring with small_lm(G.fst) and large_lm(G.carpa) .?
Failed on both win and ubuntu.
Python is 3.6 and 3.7.
See attached logs.
vosk_ubuntu.log
vosk_win.log
Like the ones from here: https://kaldi-asr.org/models.html
Also, where are those android models sourced from? And what's the effect of using a smaller model like the android ones?
Thanks!
I am randomly getting the prefix "nspc " in the recognized text. Even with German text. "nspc wann nudeln begünstigen". What is this about? :)
please share Keyword Search example with same aar
lib ?
Hi
I see armv6l version vosk-0.3.3-cp37-cp37m-linux_armv6l.whl available in PIP. Should it works on Ras Pi Zero?
I recived SIGILL, Illegal instruction when I run python example on Pi zero. It seems the library were compiled for some other system.
Gdb output:
(gdb) r
Starting program: /usr/bin/python3 test_simple.py
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/arm-linux-gnueabihf/libthread_db.so.1".
Program received signal SIGILL, Illegal instruction.
0xb65ecdd0 in fst::FstRegisterer<fst::VectorFst<fst::ArcTpl<fst::TropicalWeig
htTpl<float> >, fst::VectorState<fst::ArcTpl<fst::TropicalWeightTpl<float> >,
std::allocator<fst::ArcTpl<fst::TropicalWeightTpl<float> > > > > >::FstRegis
terer() () from /home/pi/.local/lib/python3.7/site-packages/vosk/_vosk.so
And cat /proc/cpuinfo
processor : 0
model name : ARMv6-compatible processor rev 7 (v6l)
BogoMIPS : 697.95
Features : half thumb fastmult vfp edsp java tls
CPU implementer : 0x41
CPU architecture: 7
CPU variant : 0x0
CPU part : 0xb76
CPU revision : 7
Hardware : BCM2835
Revision : 9000c1
Serial : 0000000081d02f5e
Model : Raspberry Pi Zero W Rev 1.1
Hello!
Is it possible to train our own custom models like these: https://github.com/alphacep/kaldi-android-demo/releases using Vosk? What steps shouls we take after the index database is filled with data?
Thank you!
Related to #18 , using the same code, note how the C++ exception is not handled and causes Python to abort.
Perhaps it would be possible to let Python code handle exceptions thrown from C++?
I'd like to help, but it would take me some time to learn C++ and SWIG :)
from vosk import Model, KaldiRecognizer
try:
model = Model('fr')
rec = KaldiRecognizer(model, 8000)
data = b'\x00' * 1000
print(rec.AcceptWaveform(data))
except Exception as e:
print('Exception!', e)
raise e
print('OK')
@nshmyrev this is my idea of "pythonic" api.
import vosk
# print available models
print(vosk.list_models())
# auto downloads the model if not found in local
asr = vosk.load("en-us")
# word_list is optional
# if stream=True, will return iterator for partial results
# if stream=False will return final result
# default stream = False
result = asr.recognize('wav_file_path', word_list, stream=True)
let me know if this is ok or any changes required.
Hello
First of all, thanks for this repository.
I am trying to install Vosk on Raspbian on a 4GB Raspberry Pi 4 but it cannot install version 0.3.1.
Pip (20.0.2) says that wouldn't find a version 0.3.1 and installs the 0.3. When installing whl files none of them are compatible. I am trying to avoid installing the whole Kaldi in the Raspberry since it will take a lot of free space.
I want to install this version since I cannot pass a list of words to a KaldiRecognizer in version 0.3.
Also, the model seems not to find a HCLG.fst file and requires a HCLr.fst and a Gr.fst separatedly, at least in version 0.3, when passing a model folder made by us.
Thanks for your time,
Ander
Is there an time estimation regarding the iOS solution?
Do we need to convert models trained for example using aspire, librispeech recipes to allow them to work with passing word list to lookahead
https://github.com/alphacep/vosk-api/blob/master/python/example/test_local_words.py
I have several tdnn models that I want to use with these api.
I was wondering is it possible to get nBest list from ASR result instead of only 1?
For example:
['look for movies directed by Terren Tina',
'look for movies directed by Darren Tina',
'look for movies directed by Tarantino',
'look for movies directed by tear and tea no',
'look for movies directed by Terren teen']
Thank you for a great lib!
But I didn't get a chance to try it, because I can't install python module.
There were some errors during installation:
During Kaldi installation, the command helps to install all necessary dependencies before "make -j 10".
./kaldi/tools./extras/check_dependencies.sh
I had to install CMake 3.10.1 or higher.
https://askubuntu.com/questions/829310/how-to-upgrade-cmake-in-ubuntu
I tried to run sudo python3 setup.py install
but got an error:
fatal error: Python.h: No such file or directory
I've added to CMakeLists.txt
2 lines and it worked for me.
find_package(PythonLibs REQUIRED)
include_directories(${PYTHON_INCLUDE_DIRS})
*from here https://stackoverflow.com/questions/11041299/python-h-no-such-file-or-directory
After that I installed required OpenFST as described here:
https://aghriss.github.io/posts/2018/01/01/OpenFSTubuntu.html
And a step with hack:
copied folders from kaldi/src/
to vosk-api/src/
.
I am not sure, how to do it corect, but it helped me to move further.
Now I have a lot of errors. The first is:
/usr/bin/c++
-D_vosk_EXPORTS -I/usr/include/python3.5m -I/home/smartly/tmp/vosk-api/python/../src -I/src -I/tools/openfst/include -O3 -DFST_NO_DYNAMIC_LINKING -O3 -DNDEBUG -fPIC -o CMakeFiles/_vosk.dir/voskPYTHON_wrap.cxx.o -c /home/smartly/tmp/vosk-api/python/build/temp.linux-x86_64-3.5/voskPYTHON_wrap.cxx
In file included from /usr/include/c++/5/cstdint:35:0,
from /usr/local/include/fst/types.h:19,
from /home/smartly/tmp/vosk-api/python/../src/base/kaldi-types.h:44,
from /home/smartly/tmp/vosk-api/python/../src/base/kaldi-error.h:34,
from /home/smartly/tmp/vosk-api/python/../src/base/kaldi-common.h:35,
from /home/smartly/tmp/vosk-api/python/../src/kaldi_recognizer.h:1,
from /home/smartly/tmp/vosk-api/python/build/temp.linux-x86_64-3.5/voskPYTHON_wrap.cxx:3121:
/usr/include/c++/5/bits/c++0x_warning.h:32:2: error: #error This file requires compiler and library support for the ISO C++ 2011 standard. This support must be enabled with the -std=c++11 or -std=gnu++11 compiler options.
#error This file requires compiler and library support
^
In file included from /home/smartly/tmp/vosk-api/python/../src/base/kaldi-types.h:44:0,
from /home/smartly/tmp/vosk-api/python/../src/base/kaldi-error.h:34,
from /home/smartly/tmp/vosk-api/python/../src/base/kaldi-common.h:35,
from /home/smartly/tmp/vosk-api/python/../src/kaldi_recognizer.h:1,
from /home/smartly/tmp/vosk-api/python/build/temp.linux-x86_64-3.5/voskPYTHON_wrap.cxx:3121:
/usr/local/include/fst/types.h:24:7: error: expected nested-name-specifier before ‘int8’
using int8 = int8_t;
^
/usr/local/include/fst/types.h:25:7: error: expected nested-name-specifier before ‘int16’
using int16 = int16_t;
Can you please help me to fix it?
Hello again
I try to load a model with an HCLG.fst instead of an HCLr.fst and another Gr.fst separated graphs, as happens to be in your C++ code, but the program seems not to find it in the path...
vosk --min-active=200 --max-active=3000 --beam=10.0 --lattice-beam=2.0 --acoustic-scale=1.0 --frame-subsampling-factor=3 --endpoint.silence-phones=1:2:3:4:5:6:7:8:9:10 --endpoint.rule2.min-trailing-silence=0.5 --endpoint.rule3.min-trailing-silence=1.0 --endpoint.rule4.min-trailing-silence=2.0
LOG (vosk[5.5.641-79319]:ComputeDerivedVars():ivector-extractor.cc:183) Computing derived variables for iVector extractor
LOG (vosk[5.5.641-79319]:ComputeDerivedVars():ivector-extractor.cc:204) Done.
LOG (vosk[5.5.641-79319]:RemoveOrphanNodes():nnet-nnet.cc:948) Removed 1 orphan nodes.
LOG (vosk[5.5.641-79319]:RemoveOrphanComponents():nnet-nnet.cc:847) Removing 2 orphan components.
LOG (vosk[5.5.641-79319]:Collapse():nnet-utils.cc:1472) Added 1 components, removed 2
LOG (vosk[5.5.641-79319]:CompileLooped():nnet-compile-looped.cc:345) Spent 0.146811 seconds in looped compilation.
ERROR: Fst::Read: Can't open file: model/HCLr.fst
ERROR: Fst::Read: Can't open file: model/Gr.fst
ERROR (vosk[5.5.641-79319]:KaldiRecognizer():src/kaldi_recognizer.cc:74) Can't create decoding graph
It may be my fault, but here are the files in the model folder:
Thanks for your time and quick responses,
Ander
Hi Nickolay
Is Kaldi-WebSocket support online cmvn?
https://github.com/kaldi-asr/kaldi/blob/master/egs/wsj/s5/local/chain/tuning/run_tdnn_1i.sh
https://github.com/kaldi-asr/kaldi/pull/3560/files
my results on this model are different.
best regards
I run into following issue when doing "python3 setup.py install". Kaldi indeed is cloned from lookahead-branch.
:~/vosk-api/python# python3 setup.py install running install
running bdist_egg
running egg_info
writing vosk.egg-info/PKG-INFO
writing dependency_links to vosk.egg-info/dependency_links.txt
writing top-level names to vosk.egg-info/top_level.txt
reading manifest file 'vosk.egg-info/SOURCES.txt'
writing manifest file 'vosk.egg-info/SOURCES.txt'
installing library code to build/bdist.linux-x86_64/egg
running install_lib
running build_py
running build_ext
CMake Warning at CMakeLists.txt:6 (find_package):
By not providing "FindPython.cmake" in CMAKE_MODULE_PATH this project has
asked CMake to find a package configuration file provided by "Python", but
CMake did not find one.
Could not find a package configuration file provided by "Python" with any
of the following names:
PythonConfig.cmake
python-config.cmake
Add the installation prefix of "Python" to CMAKE_PREFIX_PATH or set
"Python_DIR" to a directory containing one of the above files. If "Python"
provides a separate development package or SDK, be sure it has been
installed.
-- Configuring done
-- Generating done
-- Build files have been written to: /root/vosk-api/python/build/temp.linux-x86_64-3.6
/usr/bin/cmake -H/root/vosk-api/python -B/root/vosk-api/python/build/temp.linux-x86_64-3.6 --check-build-system CMakeFiles/Makefile.cmake 0
make -f CMakeFiles/Makefile2 _vosk
make[1]: Entering directory '/root/vosk-api/python/build/temp.linux-x86_64-3.6'
/usr/bin/cmake -H/root/vosk-api/python -B/root/vosk-api/python/build/temp.linux-x86_64-3.6 --check-build-system CMakeFiles/Makefile.cmake 0
/usr/bin/cmake -E cmake_progress_start /root/vosk-api/python/build/temp.linux-x86_64-3.6/CMakeFiles 6
make -f CMakeFiles/Makefile2 CMakeFiles/_vosk.dir/all
make[2]: Entering directory '/root/vosk-api/python/build/temp.linux-x86_64-3.6'
make -f CMakeFiles/_vosk.dir/build.make CMakeFiles/_vosk.dir/depend
make[3]: Entering directory '/root/vosk-api/python/build/temp.linux-x86_64-3.6'
cd /root/vosk-api/python/build/temp.linux-x86_64-3.6 && /usr/bin/cmake -E cmake_depends "Unix Makefiles" /root/vosk-api/python /root/vosk-api/python /root/vosk-api/python/build/temp.linux-x86_64-3.6 /root/vosk-api/python/build/temp.linux-x86_64-3.6 /root/vosk-api/python/build/temp.linux-x86_64-3.6/CMakeFiles/_vosk.dir/DependInfo.cmake --color=
make[3]: Leaving directory '/root/vosk-api/python/build/temp.linux-x86_64-3.6'
make -f CMakeFiles/_vosk.dir/build.make CMakeFiles/_vosk.dir/build
make[3]: Entering directory '/root/vosk-api/python/build/temp.linux-x86_64-3.6'
[ 16%] Building CXX object CMakeFiles/_vosk.dir/voskPYTHON_wrap.cxx.o
/usr/bin/c++ -D_vosk_EXPORTS -I/root/vosk-api/python/../src -I/root/kaldi/src -I/root/kaldi/tools/openfst/include -std=c++11 -O3 -DFST_NO_DYNAMIC_LINKING -O3 -DNDEBUG -fPIC -o CMakeFiles/_vosk.dir/voskPYTHON_wrap.cxx.o -c /root/vosk-api/python/build/temp.linux-x86_64-3.6/voskPYTHON_wrap.cxx
/root/vosk-api/python/build/temp.linux-x86_64-3.6/voskPYTHON_wrap.cxx:173:11: fatal error: Python.h: No such file or directory
# include <Python.h>
^~~~~~~~~~
compilation terminated.
CMakeFiles/_vosk.dir/build.make:71: recipe for target 'CMakeFiles/_vosk.dir/voskPYTHON_wrap.cxx.o' failed
make[3]: *** [CMakeFiles/_vosk.dir/voskPYTHON_wrap.cxx.o] Error 1
make[3]: Leaving directory '/root/vosk-api/python/build/temp.linux-x86_64-3.6'
CMakeFiles/Makefile2:67: recipe for target 'CMakeFiles/_vosk.dir/all' failed
make[2]: *** [CMakeFiles/_vosk.dir/all] Error 2
make[2]: Leaving directory '/root/vosk-api/python/build/temp.linux-x86_64-3.6'
CMakeFiles/Makefile2:79: recipe for target 'CMakeFiles/_vosk.dir/rule' failed
make[1]: *** [CMakeFiles/_vosk.dir/rule] Error 2
make[1]: Leaving directory '/root/vosk-api/python/build/temp.linux-x86_64-3.6'
Makefile:118: recipe for target '_vosk' failed
make: *** [_vosk] Error 2
Traceback (most recent call last):
File "setup.py", line 28, in <module>
python_requires='>=3.4',
File "/usr/local/lib/python3.6/dist-packages/setuptools/__init__.py", line 144, in setup
return distutils.core.setup(**attrs)
File "/usr/lib/python3.6/distutils/core.py", line 148, in setup
dist.run_commands()
File "/usr/lib/python3.6/distutils/dist.py", line 955, in run_commands
self.run_command(cmd)
File "/usr/lib/python3.6/distutils/dist.py", line 974, in run_command
cmd_obj.run()
File "/usr/local/lib/python3.6/dist-packages/setuptools/command/install.py", line 67, in run
self.do_egg_install()
File "/usr/local/lib/python3.6/dist-packages/setuptools/command/install.py", line 109, in do_egg_install
self.run_command('bdist_egg')
File "/usr/lib/python3.6/distutils/cmd.py", line 313, in run_command
self.distribution.run_command(command)
File "/usr/lib/python3.6/distutils/dist.py", line 974, in run_command
cmd_obj.run()
File "/usr/local/lib/python3.6/dist-packages/setuptools/command/bdist_egg.py", line 172, in run
cmd = self.call_command('install_lib', warn_dir=0)
File "/usr/local/lib/python3.6/dist-packages/setuptools/command/bdist_egg.py", line 158, in call_command
self.run_command(cmdname)
File "/usr/lib/python3.6/distutils/cmd.py", line 313, in run_command
self.distribution.run_command(command)
File "/usr/lib/python3.6/distutils/dist.py", line 974, in run_command
cmd_obj.run()
File "/usr/local/lib/python3.6/dist-packages/setuptools/command/install_lib.py", line 11, in run
self.build()
File "/usr/lib/python3.6/distutils/command/install_lib.py", line 107, in build
self.run_command('build_py')
File "/usr/lib/python3.6/distutils/cmd.py", line 313, in run_command
self.distribution.run_command(command)
File "/usr/lib/python3.6/distutils/dist.py", line 974, in run_command
cmd_obj.run()
File "/root/vosk-api/python/cmake.py", line 73, in run
self.run_command("build_ext")
File "/usr/lib/python3.6/distutils/cmd.py", line 313, in run_command
self.distribution.run_command(command)
File "/usr/lib/python3.6/distutils/dist.py", line 974, in run_command
cmd_obj.run()
File "/usr/local/lib/python3.6/dist-packages/setuptools/command/build_ext.py", line 87, in run
_build_ext.run(self)
File "/usr/local/lib/python3.6/dist-packages/Cython/Distutils/old_build_ext.py", line 186, in run
_build_ext.build_ext.run(self)
File "/usr/lib/python3.6/distutils/command/build_ext.py", line 339, in run
self.build_extensions()
File "/usr/local/lib/python3.6/dist-packages/Cython/Distutils/old_build_ext.py", line 194, in build_extensions
self.build_extension(ext)
File "/root/vosk-api/python/cmake.py", line 64, in build_extension
env=env)
File "/usr/lib/python3.6/subprocess.py", line 311, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['make', 'VERBOSE=1', '_vosk']' returned non-zero exit status 2.
Hi.
While trying python3 setup.py install
[ 40%] Building CXX object CMakeFiles/_vosk.dir/voskPYTHON_wrap.cxx.o
/usr/bin/c++ -D_vosk_EXPORTS -I/opt/vosk/vosk-api/python/../src -I/opt/vosk/kaldi/src -I/opt/vosk/kaldi/tools/openfst/include -I/usr/include/python3.5m -O3 -DFST_NO_DYNAMIC_LINKING -std=c++11 -O3 -DNDEBUG -fPIC -o CMakeFiles/vosk.dir/voskPYTHON_wrap.cxx.o -c /opt/vosk/vosk-api/python/build/temp.linux-x86_64-3.5/voskPYTHON_wrap.cxx
In file included from /opt/vosk/vosk-api/python/build/temp.linux-x86_64-3.5/voskPYTHON_wrap.cxx:3119:0:
/opt/vosk/vosk-api/python/../src/kaldi_recognizer.h:35:14: error: ‘LookaheadFst’ in namespace ‘fst’ does not name a template type
fst::LookaheadFst<fst::StdArc, int32> *decode_fst;
^~~~~~~~~~~~
CMakeFiles/_vosk.dir/build.make:62: recipe for target 'CMakeFiles/_vosk.dir/voskPYTHON_wrap.cxx.o' failed
make[3]: *** [CMakeFiles/_vosk.dir/voskPYTHON_wrap.cxx.o] Error 1
make[3]: Leaving directory '/opt/vosk/vosk-api/python/build/temp.linux-x86_64-3.5'
CMakeFiles/Makefile2:104: recipe for target 'CMakeFiles/_vosk.dir/all' failed
make[2]: *** [CMakeFiles/_vosk.dir/all] Error 2
make[2]: Leaving directory '/opt/vosk/vosk-api/python/build/temp.linux-x86_64-3.5'
CMakeFiles/Makefile2:116: recipe for target 'CMakeFiles/_vosk.dir/rule' failed
make[1]: *** [CMakeFiles/_vosk.dir/rule] Error 2
make[1]: Leaving directory '/opt/vosk/vosk-api/python/build/temp.linux-x86_64-3.5'
Makefile:131: recipe for target '_vosk' failed
make: *** [_vosk] Error 2
Some audio files repeat the final result
{"partial" : ""}
{"partial" : ""}
{"partial" : ""}
{"partial" : ""}
{"partial" : "are you"}
{"partial" : "i usually"}
{"partial" : "i usually get"}
{"partial" : "i usually get up"}
{"partial" : "i usually get up at"}
{"partial" : "i usually get up at a turk leave"}
{"partial" : "i usually get up at eight o'clock"}
{"partial" : "i usually get up at eight o'clock"}
{"result" : [ {"word": "i", "start" : 1.59, "end" : 1.83, "conf" : 1},
{"word": "usually", "start" : 1.83, "end" : 2.55, "conf" : 1},
{"word": "get", "start" : 2.64, "end" : 2.94, "conf" : 1},
{"word": "up", "start" : 2.94, "end" : 3.27, "conf" : 1},
{"word": "at", "start" : 3.51, "end" : 3.87, "conf" : 1},
{"word": "eight", "start" : 3.96, "end" : 4.28829, "conf" : 0.585446},
{"word": "o'clock", "start" : 4.29, "end" : 5.04, "conf" : 0.574342}
], "text" : "i usually get up at eight o'clock" }
{"result" : [ {"word": "i", "start" : 1.59, "end" : 1.83, "conf" : 1},
{"word": "usually", "start" : 1.83, "end" : 2.55, "conf" : 1},
{"word": "get", "start" : 2.64, "end" : 2.94, "conf" : 1},
{"word": "up", "start" : 2.94, "end" : 3.27, "conf" : 1},
{"word": "at", "start" : 3.51, "end" : 3.87, "conf" : 1},
{"word": "eight", "start" : 3.96, "end" : 4.28829, "conf" : 0.585446},
{"word": "o'clock", "start" : 4.29, "end" : 5.04, "conf" : 0.574342}
], "text" : "i usually get up at eight o'clock" }
Do you know what could cause this?
hi
I deployed the code on windows and used visual studio compile. My openfst is referencing the url :https://github.com/kkm000/openfst.
But I get the following error when I compile model.cc. Can you help me how to resolve it?
Many thanks.
model.obj: error LNK2019: Unresolved external symbol "public: unsigned __int64 __cdecl fst :: BitmapIndex :: Rank1 (unsigned __int64) const" (? Rank1 @ BitmapIndex @ fst @@ QEBA_K_K @ Z) in the function "public: class fst :: TropicalWeightTpl __cdecl fst :: internal :: NGramFstImpl <struct fst :: ArcTpl <class fst :: TropicalWeightTpl >> :: Final (int) const "(? Final @? $ NGramFstImpl @ U ? $ ArcTpl @ V? $ TropicalWeightTpl @ M @ fst @@@ fst @@@ internal @ fst @@ QEBA? AV? $ TropicalWeightTpl @ M @ 3 @ H @ Z) is referenced
model.obj: error LNK2019: unresolved external symbol "public: unsigned __int64 __cdecl fst :: BitmapIndex :: Select1 (unsigned __int64) const" (? Select1 @ BitmapIndex @ fst @@ QEBA_K_K @ Z) in function "public: virtual bool __cdecl fst :: NGramFstMatcher <struct fst :: ArcTpl <class fst :: TropicalWeightTpl >> :: Find (int) "(? Find @? $ NGramFstMatcher @ U? $ ArcTpl @ V? $ TropicalWeightTpl @ M @fst @@@ fst @@@ fst @@ UEAA_NH @ Z)
model.obj: error LNK2019: Unresolved external symbol "public: struct std :: pair <unsigned __int64, unsigned __int64> __cdecl fst :: BitmapIndex :: Select0s (unsigned __int64) const" (? Select0s @ BitmapIndex @ fst @@ QEBA? AU? $ Pair @ _K_K @ std @@ _ K @ Z) in the function "public: void __cdecl fst :: internal :: NGramFstImpl <struct fst :: ArcTpl <class fst :: TropicalWeightTpl > :: Init (char const *, bool, class fst :: MappedFile *) "(? Init @? $ NGramFstImpl @ U? $ ArcTpl @ V? $ TropicalWeightTpl @ M @ fst @@ fst @@@ internal @ fst @@ QEAAXPEBD_NPEAVMappedFile @ 3 @@ Z) is referenced
model.obj: error LNK2019: Unresolved external symbol "public: void __cdecl fst :: BitmapIndex :: BuildIndex (unsigned __int64 const *, unsigned __int64)" (? BuildIndex @ BitmapIndex @ fst @@ QEAAXPEB_K_K @ Z) in function "public: void __cdecl fst :: internal :: NGramFstImpl <struct fst :: ArcTpl <class fst :: TropicalWeightTpl >> :: Init (char const *, bool, class fst :: MappedFile *)" (? Init @? $ NGramFstImpl @ U? $ ArcTpl @ V? $ TropicalWeightTpl @ M @ fst @@@ fst @@@ internal @ fst @@ QEAAXPEBD_NPEAVMappedFile @ 3 @@ Z) is referenced
MSVCRTD.lib (exe_main.obj): error LNK2019: Unresolved external symbol main is referenced in function "int __cdecl invoke_main (void)" (? Invoke_main @@ YAHXZ)
C: \ Users \ celine \ kaldi \ kaldiwin_vs2017_OPENBLAS \ zzz_vosk_api \ x64 \ Debug \ zzz_vosk_api.exe: fatal error LNK1120: 5 unresolvable external symbols
请问基于什么数据库训练的模型,是中文的语音识别模型么?
Help is very much needed on all this, any contribution will be appreciated.
I think it would be good to give users options for which models they can use along with the pros and cons of each. So add the models from https://github.com/alphacep/vosk-server.
Also, it would probably be good to document the model format and how to create your own or convert one of the official kaldi models.
edit: Just noticed https://github.com/alphacep/vosk-api/blob/master/doc/model.md which is related.
I'm having an error similar to #17 here. I've compiled kaldi from git and try to build the python module. I don't do that with python3 setup.py install
, because that needs super-user rights on my system and will fail because KALDI_ROOT is only exported for the current user. I do python3 setup.py build
. But get the error:
[ 80%] Building CXX object CMakeFiles/_vosk.dir/home/magic/vosk-api/src/model.cc.o
/usr/bin/c++ -D_vosk_EXPORTS -I/home/magic/vosk-api/python/../src -I/home/magic/kaldi/src -I/home/magic/kaldi/tools/openfst/include -I/usr/include/python3.7m -O3 -DFST_NO_DYNAMIC_LINKING -O3 -DNDEBUG -fPIC -o CMakeFiles/_vosk.dir/home/magic/vosk-api/src/model.cc.o -c /home/magic/vosk-api/src/model.cc
make[3]: *** Keine Regel vorhanden, um das Ziel „/home/magic/kaldi/src/online2/kaldi-online2.a“,
benötigt von „../lib.linux-x86_64-3.7/vosk/_vosk.so“, zu erstellen. Schluss.
Any hints on what the error is and hints on installing afterwards?
Board: Orange Pi zero.
OS: Ubuntu Bionic with Armbian Linux.
# uname -a
Linux orangepizero 4.19.62-sunxi #5.92 SMP Wed Jul 31 22:07:23 CEST 2019 armv7l armv7l armv7l GNU/Linux
# python3.7 -V
Python 3.7.5
# apt-cache policy libgfortran5 | grep Installed
Installed: 8.3.0-6ubuntu1~18.04.1
vosk installed from pypi, result:
# python3.7 test_local.py test.wav
vosk --min-active=200 --max-active=3000 --beam=10.0 --lattice-beam=2.0 --acoustic-scale=1.0 --frame-subsampling-factor=3 --endpoint.silence-phones=1:2:3:4:5:6:7:8:9:10 --endpoint.rule2.min-trailing-silence=0.5 --endpoint.rule3.min-trailing-silence=1.0 --endpoint.rule4.min-trailing-silence=2.0
LOG (vosk[5.5.556~1-5ae69]:ComputeDerivedVars():ivector-extractor.cc:183) Computing derived variables for iVector extractor
LOG (vosk[5.5.556~1-5ae69]:ComputeDerivedVars():ivector-extractor.cc:204) Done.
LOG (vosk[5.5.556~1-5ae69]:RemoveOrphanNodes():nnet-nnet.cc:948) Removed 1 orphan nodes.
LOG (vosk[5.5.556~1-5ae69]:RemoveOrphanComponents():nnet-nnet.cc:847) Removing 2 orphan components.
LOG (vosk[5.5.556~1-5ae69]:Collapse():nnet-utils.cc:1472) Added 1 components, removed 2
LOG (vosk[5.5.556~1-5ae69]:CompileLooped():nnet-compile-looped.cc:345) Spent 0.521 seconds in looped compilation.
Illegal instruction
Always crashed on rec = KaldiRecognizer(model, 16000)
.
P.S.: x86_64 and self-builded aarch64 wheel work fine.
Android with the small one. It should be configurable somehow.
Right now servers have suboptimal accuracy due to the small beam.
See this pulll request. Make sure you mention it in the README.
when I run "gradle build", I got the error message:
> Task :swig FAILED
Unable to open file build/generated-src/cpp/kaldi_recognizer_wrap.cc: No such file or directory
FAILURE: Build failed with an exception.
* What went wrong:
Execution failed for task ':swig'.
> Process 'command 'swig'' finished with non-zero exit value 1
I installed swig from https://github.com/swig/swig.git. Is there anything else I missed? Thanks.
Is there any time-planning for python builds to support ARM-architectures?
Ubuntu 16.04, installed with "pip3 install vosk"
python3.5 test-vosk.py test.wav
vosk --min-active=200 --max-active=3000 --beam=10.0 --lattice-beam=2.0 --acoustic-scale=1.0 --frame-subsampling-factor=3 --endpoint.silence-phones=1:2:3:4:5:6:7:8:9:10 --endpoint.rule2.min-trailing-silence=0.5 --endpoint.rule3.min-trailing-silence=1.0 --endpoint.rule4.min-trailing-silence=2.0
ERROR (vosk[5.5.599~1-11bed]:ReadConfigFile():parse-options.cc:462) Cannot open config file: model/mfcc.conf
I have a demo of the WebAssembly version available here:
https://dtreskunov.github.io/tiny-kaldi/
Works pretty well, actually. Also, check out how small the binary is!
To reduce package size
On a Raspberry Pi Zero W with Raspbian 10 buster.
sudo apt-get install libgfortran-6-dev
pip3 install --user vosk
cd vosk-api/python/example
wget https://github.com/alphacep/kaldi-android-demo/releases/download/2020-01/alphacep-model-android-en-us-0.3.tar.gz
tar xf alphacep-model-android-en-us-0.3.tar.gz
mv alphacep-model-android-en-us-0.3 model-en
python3 ./test_simple.py test.wav
Illegal instruction
I am following this blog(https://medium.com/swlh/compile-kaldi-for-64-bit-android-on-ubuntu-18-70967eb3a308) to build kaldi for android. In the step Compile libkaldi_jni.so
, this blog used this repo to build libkaldi_jni.so
using the gradle build
command. However, I am wondering if anyone can help out, the following error message showed up:
In file included from /home/liuyi/Documents/VoiceAssist/kaldi-jni/build/generated-src/cpp/kaldi_recognizer_wrap.cc:236:
../../../../src/main/cpp/kaldi_recognizer.h:35:14: error: no template named 'LookaheadFst' in namespace 'fst'
fst::LookaheadFst<fst::StdArc, int32> *decode_fst_;
~~~~~^
1 error generated.
>
Task :externalNativeBuildDebug
Build kaldi_jni_arm64-v8a
[1/4] /home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/bin/clang++ --target=aarch64-none-linux-android21 --gcc-toolchain=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64 --sysroot=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/sysroot -Dkaldi_jni_EXPORTS -I../../../../src/main/cpp -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/src -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/tools/openfst-1.6.8/include -g -DANDROID -fdata-sections -ffunction-sections -funwind-tables -fstack-protector-strong -no-canonical-prefixes -fno-addrsig -Wa,--noexecstack -Wformat -Werror=format-security -O3 -DFST_NO_DYNAMIC_LINKING -O0 -fno-limit-debug-info -fno-experimental-isel -fPIC -MD -MT CMakeFiles/kaldi_jni.dir/build/generated-src/cpp/kaldi_recognizer_wrap.cc.o -MF CMakeFiles/kaldi_jni.dir/build/generated-src/cpp/kaldi_recognizer_wrap.cc.o.d -o CMakeFiles/kaldi_jni.dir/build/generated-src/cpp/kaldi_recognizer_wrap.cc.o -c /home/liuyi/Documents/VoiceAssist/kaldi-jni/build/generated-src/cpp/kaldi_recognizer_wrap.cc
[2/4] /home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/bin/clang++ --target=aarch64-none-linux-android21 --gcc-toolchain=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64 --sysroot=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/sysroot -Dkaldi_jni_EXPORTS -I../../../../src/main/cpp -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/src -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/tools/openfst-1.6.8/include -g -DANDROID -fdata-sections -ffunction-sections -funwind-tables -fstack-protector-strong -no-canonical-prefixes -fno-addrsig -Wa,--noexecstack -Wformat -Werror=format-security -O3 -DFST_NO_DYNAMIC_LINKING -O0 -fno-limit-debug-info -fno-experimental-isel -fPIC -MD -MT CMakeFiles/kaldi_jni.dir/src/main/cpp/kaldi_recognizer.cc.o -MF CMakeFiles/kaldi_jni.dir/src/main/cpp/kaldi_recognizer.cc.o.d -o CMakeFiles/kaldi_jni.dir/src/main/cpp/kaldi_recognizer.cc.o -c /home/liuyi/Documents/VoiceAssist/kaldi-jni/src/main/cpp/kaldi_recognizer.cc
[3/4] /home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/bin/clang++ --target=aarch64-none-linux-android21 --gcc-toolchain=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64 --sysroot=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/sysroot -Dkaldi_jni_EXPORTS -I../../../../src/main/cpp -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/src -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/tools/openfst-1.6.8/include -g -DANDROID -fdata-sections -ffunction-sections -funwind-tables -fstack-protector-strong -no-canonical-prefixes -fno-addrsig -Wa,--noexecstack -Wformat -Werror=format-security -O3 -DFST_NO_DYNAMIC_LINKING -O0 -fno-limit-debug-info -fno-experimental-isel -fPIC -MD -MT CMakeFiles/kaldi_jni.dir/src/main/cpp/model.cc.o -MF CMakeFiles/kaldi_jni.dir/src/main/cpp/model.cc.o.d -o CMakeFiles/kaldi_jni.dir/src/main/cpp/model.cc.o -c /home/liuyi/Documents/VoiceAssist/kaldi-jni/src/main/cpp/model.cc
FAILED: /home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/bin/clang++ --target=aarch64-none-linux-android21 --gcc-toolchain=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64 --sysroot=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/sysroot -Dkaldi_jni_EXPORTS -I../../../../src/main/cpp -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/src -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/tools/openfst-1.6.8/include -g -DANDROID -fdata-sections -ffunction-sections -funwind-tables -fstack-protector-strong -no-canonical-prefixes -fno-addrsig -Wa,--noexecstack -Wformat -Werror=format-security -O3 -DFST_NO_DYNAMIC_LINKING -O0 -fno-limit-debug-info -fno-experimental-isel -fPIC -MD -MT CMakeFiles/kaldi_jni.dir/build/generated-src/cpp/kaldi_recognizer_wrap.cc.o -MF CMakeFiles/kaldi_jni.dir/build/generated-src/cpp/kaldi_recognizer_wrap.cc.o.d -o CMakeFiles/kaldi_jni.dir/build/generated-src/cpp/kaldi_recognizer_wrap.cc.o -c /home/liuyi/Documents/VoiceAssist/kaldi-jni/build/generated-src/cpp/kaldi_recognizer_wrap.cc
In file included from /home/liuyi/Documents/VoiceAssist/kaldi-jni/build/generated-src/cpp/kaldi_recognizer_wrap.cc:236:
../../../../src/main/cpp/kaldi_recognizer.h:35:14: error: no template named 'LookaheadFst' in namespace 'fst'
fst::LookaheadFst<fst::StdArc, int32> *decode_fst_;
~~~~~^
1 error generated.
FAILED: /home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/bin/clang++ --target=aarch64-none-linux-android21 --gcc-toolchain=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64 --sysroot=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/sysroot -Dkaldi_jni_EXPORTS -I../../../../src/main/cpp -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/src -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/tools/openfst-1.6.8/include -g -DANDROID -fdata-sections -ffunction-sections -funwind-tables -fstack-protector-strong -no-canonical-prefixes -fno-addrsig -Wa,--noexecstack -Wformat -Werror=format-security -O3 -DFST_NO_DYNAMIC_LINKING -O0 -fno-limit-debug-info -fno-experimental-isel -fPIC -MD -MT CMakeFiles/kaldi_jni.dir/src/main/cpp/kaldi_recognizer.cc.o -MF CMakeFiles/kaldi_jni.dir/src/main/cpp/kaldi_recognizer.cc.o.d -o CMakeFiles/kaldi_jni.dir/src/main/cpp/kaldi_recognizer.cc.o -c /home/liuyi/Documents/VoiceAssist/kaldi-jni/src/main/cpp/kaldi_recognizer.cc
In file included from /home/liuyi/Documents/VoiceAssist/kaldi-jni/src/main/cpp/kaldi_recognizer.cc:1:
../../../../src/main/cpp/kaldi_recognizer.h:35:14: error: no template named 'LookaheadFst' in namespace 'fst'
fst::LookaheadFst<fst::StdArc, int32> *decode_fst_;
~~~~~^
/home/liuyi/Documents/VoiceAssist/kaldi-jni/src/main/cpp/kaldi_recognizer.cc:14:19: error: use of undeclared identifier 'LookaheadComposeFst'
decode_fst_ = LookaheadComposeFst(*model_.hcl_fst_, *model_.g_fst_, model_.disambig_);
^
2 errors generated.
ninja: build stopped: subcommand failed.
>
Task :externalNativeBuildDebug FAILED
FAILURE: Build failed with an exception.
*
What went wrong:
Execution failed for task ':externalNativeBuildDebug'.
>
Build command failed.
Error while executing process /home/liuyi/Android/Sdk/cmake/3.6.4111459/bin/cmake with arguments {--build /home/liuyi/Documents/VoiceAssist/kaldi-jni/.cxx/cmake/debug/arm64-v8a --target kaldi_jni}
[1/4] /home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/bin/clang++ --target=aarch64-none-linux-android21 --gcc-toolchain=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64 --sysroot=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/sysroot -Dkaldi_jni_EXPORTS -I../../../../src/main/cpp -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/src -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/tools/openfst-1.6.8/include -g -DANDROID -fdata-sections -ffunction-sections -funwind-tables -fstack-protector-strong -no-canonical-prefixes -fno-addrsig -Wa,--noexecstack -Wformat -Werror=format-security -O3 -DFST_NO_DYNAMIC_LINKING -O0 -fno-limit-debug-info -fno-experimental-isel -fPIC -MD -MT CMakeFiles/kaldi_jni.dir/build/generated-src/cpp/kaldi_recognizer_wrap.cc.o -MF CMakeFiles/kaldi_jni.dir/build/generated-src/cpp/kaldi_recognizer_wrap.cc.o.d -o CMakeFiles/kaldi_jni.dir/build/generated-src/cpp/kaldi_recognizer_wrap.cc.o -c /home/liuyi/Documents/VoiceAssist/kaldi-jni/build/generated-src/cpp/kaldi_recognizer_wrap.cc
[2/4] /home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/bin/clang++ --target=aarch64-none-linux-android21 --gcc-toolchain=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64 --sysroot=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/sysroot -Dkaldi_jni_EXPORTS -I../../../../src/main/cpp -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/src -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/tools/openfst-1.6.8/include -g -DANDROID -fdata-sections -ffunction-sections -funwind-tables -fstack-protector-strong -no-canonical-prefixes -fno-addrsig -Wa,--noexecstack -Wformat -Werror=format-security -O3 -DFST_NO_DYNAMIC_LINKING -O0 -fno-limit-debug-info -fno-experimental-isel -fPIC -MD -MT CMakeFiles/kaldi_jni.dir/src/main/cpp/kaldi_recognizer.cc.o -MF CMakeFiles/kaldi_jni.dir/src/main/cpp/kaldi_recognizer.cc.o.d -o CMakeFiles/kaldi_jni.dir/src/main/cpp/kaldi_recognizer.cc.o -c /home/liuyi/Documents/VoiceAssist/kaldi-jni/src/main/cpp/kaldi_recognizer.cc
[3/4] /home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/bin/clang++ --target=aarch64-none-linux-android21 --gcc-toolchain=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64 --sysroot=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/sysroot -Dkaldi_jni_EXPORTS -I../../../../src/main/cpp -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/src -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/tools/openfst-1.6.8/include -g -DANDROID -fdata-sections -ffunction-sections -funwind-tables -fstack-protector-strong -no-canonical-prefixes -fno-addrsig -Wa,--noexecstack -Wformat -Werror=format-security -O3 -DFST_NO_DYNAMIC_LINKING -O0 -fno-limit-debug-info -fno-experimental-isel -fPIC -MD -MT CMakeFiles/kaldi_jni.dir/src/main/cpp/model.cc.o -MF CMakeFiles/kaldi_jni.dir/src/main/cpp/model.cc.o.d -o CMakeFiles/kaldi_jni.dir/src/main/cpp/model.cc.o -c /home/liuyi/Documents/VoiceAssist/kaldi-jni/src/main/cpp/model.cc
FAILED: /home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/bin/clang++ --target=aarch64-none-linux-android21 --gcc-toolchain=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64 --sysroot=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/sysroot -Dkaldi_jni_EXPORTS -I../../../../src/main/cpp -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/src -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/tools/openfst-1.6.8/include -g -DANDROID -fdata-sections -ffunction-sections -funwind-tables -fstack-protector-strong -no-canonical-prefixes -fno-addrsig -Wa,--noexecstack -Wformat -Werror=format-security -O3 -DFST_NO_DYNAMIC_LINKING -O0 -fno-limit-debug-info -fno-experimental-isel -fPIC -MD -MT CMakeFiles/kaldi_jni.dir/build/generated-src/cpp/kaldi_recognizer_wrap.cc.o -MF CMakeFiles/kaldi_jni.dir/build/generated-src/cpp/kaldi_recognizer_wrap.cc.o.d -o CMakeFiles/kaldi_jni.dir/build/generated-src/cpp/kaldi_recognizer_wrap.cc.o -c /home/liuyi/Documents/VoiceAssist/kaldi-jni/build/generated-src/cpp/kaldi_recognizer_wrap.cc
In file included from /home/liuyi/Documents/VoiceAssist/kaldi-jni/build/generated-src/cpp/kaldi_recognizer_wrap.cc:236:
../../../../src/main/cpp/kaldi_recognizer.h:35:14: error: no template named 'LookaheadFst' in namespace 'fst'
fst::LookaheadFst<fst::StdArc, int32> *decode_fst_;
~~~~~^
1 error generated.
FAILED: /home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/bin/clang++ --target=aarch64-none-linux-android21 --gcc-toolchain=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64 --sysroot=/home/liuyi/Android/Sdk/ndk/20.1.5948944/toolchains/llvm/prebuilt/linux-x86_64/sysroot -Dkaldi_jni_EXPORTS -I../../../../src/main/cpp -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/src -I/home/liuyi/Documents/VoiceAssist/kaldi-android-renew/tools/openfst-1.6.8/include -g -DANDROID -fdata-sections -ffunction-sections -funwind-tables -fstack-protector-strong -no-canonical-prefixes -fno-addrsig -Wa,--noexecstack -Wformat -Werror=format-security -O3 -DFST_NO_DYNAMIC_LINKING -O0 -fno-limit-debug-info -fno-experimental-isel -fPIC -MD -MT CMakeFiles/kaldi_jni.dir/src/main/cpp/kaldi_recognizer.cc.o -MF CMakeFiles/kaldi_jni.dir/src/main/cpp/kaldi_recognizer.cc.o.d -o CMakeFiles/kaldi_jni.dir/src/main/cpp/kaldi_recognizer.cc.o -c /home/liuyi/Documents/VoiceAssist/kaldi-jni/src/main/cpp/kaldi_recognizer.cc
In file included from /home/liuyi/Documents/VoiceAssist/kaldi-jni/src/main/cpp/kaldi_recognizer.cc:1:
../../../../src/main/cpp/kaldi_recognizer.h:35:14: error: no template named 'LookaheadFst' in namespace 'fst'
fst::LookaheadFst<fst::StdArc, int32> *decode_fst_;
~~~~~^
/home/liuyi/Documents/VoiceAssist/kaldi-jni/src/main/cpp/kaldi_recognizer.cc:14:19: error: use of undeclared identifier 'LookaheadComposeFst'
decode_fst_ = LookaheadComposeFst(*model_.hcl_fst_, *model_.g_fst_, model_.disambig_);
^
2 errors generated.
ninja: build stopped: subcommand failed.
*
Try:
Run with --stacktrace option to get the stack trace. Run with --info or --debug option to get more log output. Run with --scan to get full insights.
*
Get more help at https://help.gradle.org
Deprecated Gradle features were used in this build, making it incompatible with Gradle 7.0.
Use '--warning-mode all' to show the individual deprecation warnings.
See https://docs.gradle.org/6.0/userguide/command_line_interface.html#sec:command_line_warnings
BUILD FAILED in 13s
22 actionable tasks: 3 executed, 19 up-to-date
Please let me know if there are any confusions in my explanation here.
In python/CMakeLists.txt
Line 6, seems cannot find python.h
.
find_package (Python COMPONENTS Interpreter Development)
I'm using anaconda3 for compilation.
I need to hardcode my lib and include path like this:
set(Python_INCLUDE_DIR "/opt/anaconda3/envs/python37/include/python3.7m")
set(Python_LIBRARY "/opt/anaconda3/envs/python37/lib/python3.7")
Is there a way to fix it?
Cmake version is 3.16.4 btw.
After the installation OK with OpenBLAS, errors comes when unzip and install fst. The error message is below:
Install OK!
make[1]: Leaving directory '/home/liuyi/Documents/VoiceAssist/vosk-api/android/build/kaldi_arm_64/OpenBLAS'
make: Leaving directory '/home/liuyi/Documents/VoiceAssist/vosk-api/android/build/kaldi_arm_64/OpenBLAS'
cd /home/liuyi/Documents/VoiceAssist/vosk-api/android/build/kaldi_arm_64
wget -c -T 10 -t 1 http://www.openfst.org/twiki/pub/FST/FstDownload/openfst-1.6.7.tar.gz
--2020-01-03 15:00:45-- http://www.openfst.org/twiki/pub/FST/FstDownload/openfst-1.6.7.tar.gz
Resolving www.openfst.org (www.openfst.org)... 206.196.111.47
Connecting to www.openfst.org (www.openfst.org)|206.196.111.47|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 1230836 (1.2M) [application/x-gzip]
Saving to: ‘openfst-1.6.7.tar.gz’0K .......... .......... .......... .......... .......... 4% 571K 2s
50K .......... .......... .......... .......... .......... 8% 1010K 2s
100K .......... .......... .......... .......... .......... 12% 11.5M 1s
150K .......... .......... .......... .......... .......... 16% 865K 1s
200K .......... .......... .......... .......... .......... 20% 10.9M 1s
250K .......... .......... .......... .......... .......... 24% 11.6M 1s
300K .......... .......... .......... .......... .......... 29% 25.1K 5s
350K .......... .......... .......... .......... .......... 33% 163M 4s
400K ...... 33% 413 =17s2020-01-03 15:01:13 (23.6 KB/s) - Read error at byte 415799/1230836 (Connection timed out). Giving up.
wget -c -T 10 -t 3 http://www.openslr.org/resources/2/openfst-1.6.7.tar.gz
--2020-01-03 15:01:13-- http://www.openslr.org/resources/2/openfst-1.6.7.tar.gz
Resolving www.openslr.org (www.openslr.org)... 46.101.158.64
Connecting to www.openslr.org (www.openslr.org)|46.101.158.64|:80... connected.
HTTP request sent, awaiting response... 416 Requested Range Not SatisfiableThe file is already fully retrieved; nothing to do.
tar -zxvf openfst-1.6.7.tar.gz
openfst-1.6.7/
openfst-1.6.7/ltmain.sh
openfst-1.6.7/install-sh
openfst-1.6.7/aclocal.m4
openfst-1.6.7/missing
openfst-1.6.7/ar-lib
openfst-1.6.7/INSTALL
openfst-1.6.7/COPYING
openfst-1.6.7/README
openfst-1.6.7/m4/
openfst-1.6.7/m4/ltversion.m4
openfst-1.6.7/m4/libtool.m4
openfst-1.6.7/m4/ltoptions.m4
openfst-1.6.7/m4/lt~obsolete.m4
openfst-1.6.7/m4/ltsugar.m4
openfst-1.6.7/m4/ac_python_devel.m4
openfst-1.6.7/compile
openfst-1.6.7/depcomp
openfst-1.6.7/AUTHORS
openfst-1.6.7/NEWS
openfst-1.6.7/Makefile.in
openfst-1.6.7/config.h.in
openfst-1.6.7/configure.ac
openfst-1.6.7/config.guess
openfst-1.6.7/test-driver
openfst-1.6.7/src/
openfst-1.6.7/src/bin/
openfst-1.6.7/src/bin/fstshortestpath-main.cc
openfst-1.6.7/src/bin/fstprint.cc
openfst-1.6.7/src/bin/fstcompile-main.cc
openfst-1.6.7/src/bin/fstisomorphic.cc
openfst-1.6.7/src/bin/fstcompose-main.cc
openfst-1.6.7/src/bin/fstconnect-main.cc
openfst-1.6.7/src/bin/fstprune.cc
openfst-1.6.7/src/bin/fstinfo.cc
openfst-1.6.7/src/bin/fstdraw.cc
openfst-1.6.7/src/bin/fstprint-main.cc
openfst-1.6.7/src/bin/fstunion.cc
openfst-1.6.7/src/bin/fstrandgen-main.cc
openfst-1.6.7/src/bin/fstrandgen.cc
openfst-1.6.7/src/bin/fstsymbols-main.cc
openfst-1.6.7/src/bin/fstepsnormalize-main.cc
openfst-1.6.7/src/bin/fstpush.cc
openfst-1.6.7/src/bin/fstconcat-main.cc
openfst-1.6.7/src/bin/fstdeterminize-main.cc
openfst-1.6.7/src/bin/fstencode-main.cc
openfst-1.6.7/src/bin/fstmap.cc
openfst-1.6.7/src/bin/fstisomorphic-main.cc
openfst-1.6.7/src/bin/fstminimize-main.cc
openfst-1.6.7/src/bin/fstreweight-main.cc
openfst-1.6.7/src/bin/fstreplace-main.cc
openfst-1.6.7/src/bin/fstarcsort-main.cc
openfst-1.6.7/src/bin/fstdifference-main.cc
openfst-1.6.7/src/bin/fstpush-main.cc
openfst-1.6.7/src/bin/Makefile.in
openfst-1.6.7/src/bin/fstsymbols.cc
openfst-1.6.7/src/bin/fstepsnormalize.cc
openfst-1.6.7/src/bin/fstconvert.cc
openfst-1.6.7/src/bin/fstintersect-main.cc
openfst-1.6.7/src/bin/fstshortestpath.cc
openfst-1.6.7/src/bin/fstdifference.cc
openfst-1.6.7/src/bin/fstconnect.cc
openfst-1.6.7/src/bin/fstinvert-main.cc
openfst-1.6.7/src/bin/fstequivalent-main.cc
openfst-1.6.7/src/bin/fstequal-main.cc
openfst-1.6.7/src/bin/fstclosure.cc
openfst-1.6.7/src/bin/fstunion-main.cc
openfst-1.6.7/src/bin/fstencode.cc
openfst-1.6.7/src/bin/fstcompose.cc
openfst-1.6.7/src/bin/fstprune-main.cc
openfst-1.6.7/src/bin/fstminimize.cc
openfst-1.6.7/src/bin/fsttopsort.cc
openfst-1.6.7/src/bin/fstinvert.cc
openfst-1.6.7/src/bin/fstproject-main.cc
openfst-1.6.7/src/bin/fstintersect.cc
openfst-1.6.7/src/bin/fstinfo-main.cc
openfst-1.6.7/src/bin/fstreweight.cc
openfst-1.6.7/src/bin/fstclosure-main.cc
openfst-1.6.7/src/bin/fsttopsort-main.cc
openfst-1.6.7/src/bin/fstcompile.cc
openfst-1.6.7/src/bin/fstreplace.cc
openfst-1.6.7/src/bin/fstconcat.cc
openfst-1.6.7/src/bin/fstdisambiguate-main.cc
openfst-1.6.7/src/bin/fstreverse-main.cc
openfst-1.6.7/src/bin/fstproject.cc
openfst-1.6.7/src/bin/fstsynchronize-main.cc
openfst-1.6.7/src/bin/fstequivalent.cc
openfst-1.6.7/src/bin/fstdraw-main.cc
openfst-1.6.7/src/bin/fstrelabel.cc
openfst-1.6.7/src/bin/fstarcsort.cc
openfst-1.6.7/src/bin/fstdeterminize.cc
openfst-1.6.7/src/bin/Makefile.am
openfst-1.6.7/src/bin/fstconvert-main.cc
openfst-1.6.7/src/bin/fstrelabel-main.cc
openfst-1.6.7/src/bin/fstreverse.cc
openfst-1.6.7/src/bin/fstdisambiguate.cc
openfst-1.6.7/src/bin/fstshortestdistance.cc
openfst-1.6.7/src/bin/fstshortestdistance-main.cc
openfst-1.6.7/src/bin/fstequal.cc
openfst-1.6.7/src/bin/fstrmepsilon.cc
openfst-1.6.7/src/bin/fstsynchronize.cc
openfst-1.6.7/src/bin/fstmap-main.cc
openfst-1.6.7/src/bin/fstrmepsilon-main.cc
openfst-1.6.7/src/Makefile.in
openfst-1.6.7/src/include/
openfst-1.6.7/src/include/Makefile.in
openfst-1.6.7/src/include/Makefile.am
openfst-1.6.7/src/include/fst/
openfst-1.6.7/src/include/fst/reweight.h
openfst-1.6.7/src/include/fst/union.h
openfst-1.6.7/src/include/fst/replace-util.h
openfst-1.6.7/src/include/fst/visit.h
openfst-1.6.7/src/include/fst/pair-weight.h
openfst-1.6.7/src/include/fst/generic-register.h
openfst-1.6.7/src/include/fst/flags.h
openfst-1.6.7/src/include/fst/weight.h
openfst-1.6.7/src/include/fst/compose.h
openfst-1.6.7/src/include/fst/randequivalent.h
openfst-1.6.7/src/include/fst/expanded-fst.h
openfst-1.6.7/src/include/fst/set-weight.h
openfst-1.6.7/src/include/fst/arc-map.h
openfst-1.6.7/src/include/fst/determinize.h
openfst-1.6.7/src/include/fst/types.h
openfst-1.6.7/src/include/fst/config.h
openfst-1.6.7/src/include/fst/lookahead-matcher.h
openfst-1.6.7/src/include/fst/statesort.h
openfst-1.6.7/src/include/fst/prune.h
openfst-1.6.7/src/include/fst/verify.h
openfst-1.6.7/src/include/fst/rmfinalepsilon.h
openfst-1.6.7/src/include/fst/register.h
openfst-1.6.7/src/include/fst/state-map.h
openfst-1.6.7/src/include/fst/state-table.h
openfst-1.6.7/src/include/fst/properties.h
openfst-1.6.7/src/include/fst/replace.h
openfst-1.6.7/src/include/fst/edit-fst.h
openfst-1.6.7/src/include/fst/map.h
openfst-1.6.7/src/include/fst/arcfilter.h
openfst-1.6.7/src/include/fst/matcher.h
openfst-1.6.7/src/include/fst/union-weight.h
openfst-1.6.7/src/include/fst/product-weight.h
openfst-1.6.7/src/include/fst/label-reachable.h
openfst-1.6.7/src/include/fst/icu.h
openfst-1.6.7/src/include/fst/symbol-table.h
openfst-1.6.7/src/include/fst/const-fst.h
openfst-1.6.7/src/include/fst/equivalent.h
openfst-1.6.7/src/include/fst/util.h
openfst-1.6.7/src/include/fst/filter-state.h
openfst-1.6.7/src/include/fst/union-find.h
openfst-1.6.7/src/include/fst/complement.h
openfst-1.6.7/src/include/fst/dfs-visit.h
openfst-1.6.7/src/include/fst/relabel.h
openfst-1.6.7/src/include/fst/sparse-power-weight.h
openfst-1.6.7/src/include/fst/float-weight.h
openfst-1.6.7/src/include/fst/epsnormalize.h
openfst-1.6.7/src/include/fst/push.h
openfst-1.6.7/src/include/fst/rational.h
openfst-1.6.7/src/include/fst/config.h.in
openfst-1.6.7/src/include/fst/compact-fst.h
openfst-1.6.7/src/include/fst/fst-decl.h
openfst-1.6.7/src/include/fst/symbol-table-ops.h
openfst-1.6.7/src/include/fst/lock.h
openfst-1.6.7/src/include/fst/mapped-file.h
openfst-1.6.7/src/include/fst/matcher-fst.h
openfst-1.6.7/src/include/fst/arc-arena.h
openfst-1.6.7/src/include/fst/synchronize.h
openfst-1.6.7/src/include/fst/project.h
openfst-1.6.7/src/include/fst/state-reachable.h
openfst-1.6.7/src/include/fst/vector-fst.h
openfst-1.6.7/src/include/fst/sparse-tuple-weight.h
openfst-1.6.7/src/include/fst/heap.h
openfst-1.6.7/src/include/fst/memory.hgzip: stdin: unexpected end of file
tar: Unexpected EOF in archive
tar: Unexpected EOF in archive
tar: Error is not recoverable: exiting now
- cd openfst-1.6.7
- CXX=aarch64-linux-android21-clang++
- CXXFLAGS=' -O3 -DFST_NO_DYNAMIC_LINKING'
- ./configure --prefix=/home/liuyi/Documents/VoiceAssist/vosk-api/android/build/kaldi_arm_64/local --enable-shared --enable-static --with-pic --disable-bin --enable-lookahead-fsts --enable-ngram-fsts --host=aarch64-linux-android --build=x86-linux-gnu
./build-kaldi.sh: line 82: ./configure: No such file or directory- make -j 8
make: *** No targets specified and no makefile found. Stop.- make install
make: *** No rule to make target 'install'. Stop.- cd /home/liuyi/Documents/VoiceAssist/vosk-api/android/build/kaldi_arm_64
- git clone -b android-mix --single-branch https://github.com/alphacep/kaldi
Cloning into 'kaldi'...- cd /home/liuyi/Documents/VoiceAssist/vosk-api/android/build/kaldi_arm_64/kaldi/src
- CXX=aarch64-linux-android21-clang++
- CXXFLAGS=' -O3 -DFST_NO_DYNAMIC_LINKING'
- ./configure --use-cuda=no --mathlib=OPENBLAS --shared --android-incdir=/home/liuyi/android/sdk/ndk-bundle/toolchains/llvm/prebuilt/linux-x86_64/sysroot/usr/include --host=aarch64-linux-android --openblas-root=/home/liuyi/Documents/VoiceAssist/vosk-api/android/build/kaldi_arm_64/local --fst-root=/home/liuyi/Documents/VoiceAssist/vosk-api/android/build/kaldi_arm_64/local --fst-version=1.6.7
Configuring KALDI to use OPENBLAS.
Checking compiler aarch64-linux-android21-clang++ ...
Checking OpenFst library in /home/liuyi/Documents/VoiceAssist/vosk-api/android/build/kaldi_arm_64/local ...
***configure failed: Could not find file /home/liuyi/Documents/VoiceAssist/vosk-api/android/build/kaldi_arm_64/local/include/fst/fst.h:
you may not have installed OpenFst. See ../tools/INSTALL ***- make -j 8 depend
Makefile:30: kaldi.mk: No such file or directory
kaldi.mk does not exist; you have to run ./configure
Makefile:90: recipe for target 'kaldi.mk' failed
make: *** [kaldi.mk] Error 1- make -j 8 online2
Makefile:30: kaldi.mk: No such file or directory
kaldi.mk does not exist; you have to run ./configure
Makefile:90: recipe for target 'kaldi.mk' failed
make: *** [kaldi.mk] Error 1Task :kaldi FAILED
FAILURE: Build failed with an exception.
What went wrong:
Execution failed for task ':kaldi'.
Process 'command './build-kaldi.sh'' finished with non-zero exit value 2Try:
Run with --stacktrace option to get the stack trace. Run with --info or --debug option to get more log output. Run with --scan to get full insights.Get more help at https://help.gradle.org
Deprecated Gradle features were used in this build, making it incompatible with Gradle 7.0.
Use '--warning-mode all' to show the individual deprecation warnings.
See https://docs.gradle.org/6.0/userguide/command_line_interface.html#sec:command_line_warningsBUILD FAILED in 8m 57s
1 actionable task: 1 executed
The model directory containing HCLr.fst and Gr.fst can you please show me how to produce them? I found a script to produce HCL.fst and G.fst at https://github.com/opendcd/opendcd/blob/master/script/makeclevel.sh. Can i use it?
Thanks!
CMakeLIsts.txt need the file named kaldi_recognizer_wrap.cc ,but i cant find it.
add_library( kaldi_jni SHARED
build/generated-src/cpp/kaldi_recognizer_wrap.cc
${API_SOURCES}
)
I was able to get the lookahead model working, but it seems to recognize only words from the wordlist, for example, lets say I pass "one two three" and the person speaks my name is alpha and I can count to one two three, it produces wrong results because the words prior to one two three are also recognized as one two three. I wanted to implement a phrase hint system for example, perform normal asr but put more weight on words passed through a list.
I just tried to build the android module and got an error when generating the jni file gradle build:
/vosk-api/android/build/generated-src/cpp/vosk_wrap.cc:439: error: undefined reference to 'SpkModel::SpkModel(char const*)' clang++: error: linker command failed with exit code 1 (use -v to see invocation) ninja: build stopped: subcommand failed.
is this a swig problem finding the spk_model.h in the root src folder?
Hi,
I cant seem to find 'LookaheadComposeFst', I am trying to build your wrapper classes on mac.
Any pointers as to what file I need to include?
I tried the android build. gradle build
does not work. The error message is as follows:
(base) liuyi@liuyi-HP:android$ gradle build
Starting a Gradle Daemon (subsequent builds will be faster)
Task :lint FAILED
FAILURE: Build failed with an exception.
What went wrong:
A problem was found with the configuration of task ':lint' (type 'LintGlobalTask').
No value has been specified for property 'lintClassPath'.
Try:
Run with --stacktrace option to get the stack trace. Run with --info or --debug option to get more log output. Run with --scan to get full insights.
Get more help at https://help.gradle.org
Deprecated Gradle features were used in this build, making it incompatible with Gradle 7.0.
Use '--warning-mode all' to show the individual deprecation warnings.
See https://docs.gradle.org/6.0/userguide/command_line_interface.html#sec:command_line_warnings
BUILD FAILED in 5s
1 actionable task: 1 executed
Peppermint Linux 10
stefan@stefan-HP-255-G7-Notebook-PC ~/dev/vosk-api-master/java $ KALDI_ROOT=/home/stefan/dev/kaldi-master make
g++ -shared -o libvosk_jni.so -I/home/stefan/dev/jdk-13/include -I/home/stefan/dev/jdk-13/include/linux -I/home/stefan/dev/kaldi-master/src -I/home/stefan/dev/kaldi-master/tools/openfst/include -I../src -g -O2 -DPIC -fPIC -Wno-unused-function vosk_wrap.cc ../src/kaldi_recognizer.cc ../src/kaldi_recognizer.h ../src/model.cc ../src/model.h /home/stefan/dev/kaldi-master/src/online2/kaldi-online2.a /home/stefan/dev/kaldi-master/src/decoder/kaldi-decoder.a /home/stefan/dev/kaldi-master/src/ivector/kaldi-ivector.a /home/stefan/dev/kaldi-master/src/gmm/kaldi-gmm.a /home/stefan/dev/kaldi-master/src/nnet3/kaldi-nnet3.a /home/stefan/dev/kaldi-master/src/tree/kaldi-tree.a /home/stefan/dev/kaldi-master/src/feat/kaldi-feat.a /home/stefan/dev/kaldi-master/src/lat/kaldi-lat.a /home/stefan/dev/kaldi-master/src/hmm/kaldi-hmm.a /home/stefan/dev/kaldi-master/src/transform/kaldi-transform.a /home/stefan/dev/kaldi-master/src/cudamatrix/kaldi-cudamatrix.a /home/stefan/dev/kaldi-master/src/matrix/kaldi-matrix.a /home/stefan/dev/kaldi-master/src/fstext/kaldi-fstext.a /home/stefan/dev/kaldi-master/src/util/kaldi-util.a /home/stefan/dev/kaldi-master/src/base/kaldi-base.a /home/stefan/dev/kaldi-master/tools/openfst/lib/libfst.a /home/stefan/dev/kaldi-master/tools/openfst/lib/libfstngram.a /home/stefan/dev/kaldi-master/tools/OpenBLAS/libopenblas.a -lgfortran
In file included from vosk_wrap.cc:236:0:
../src/kaldi_recognizer.h:50:14: error: ‘LookaheadFst’ in namespace ‘fst’ does not name a template type
fst::LookaheadFst<fst::StdArc, int32> *decode_fst_;
^~~~~~~~~~~~
In file included from ../src/kaldi_recognizer.cc:15:0:
../src/kaldi_recognizer.h:50:14: error: ‘LookaheadFst’ in namespace ‘fst’ does not name a template type
fst::LookaheadFst<fst::StdArc, int32> *decode_fst_;
^~~~~~~~~~~~
../src/kaldi_recognizer.cc: In constructor ‘KaldiRecognizer::KaldiRecognizer(Model&, float)’:
../src/kaldi_recognizer.cc:28:5: error: ‘decode_fst_’ was not declared in this scope
decode_fst_ = NULL;
^~~~~~~~~~~
../src/kaldi_recognizer.cc:28:5: note: suggested alternative: ‘decoder_’
decode_fst_ = NULL;
^~~~~~~~~~~
decoder_
../src/kaldi_recognizer.cc:32:27: error: ‘LookaheadComposeFst’ was not declared in this scope
decode_fst_ = LookaheadComposeFst(*model_.hcl_fst_, *model_.g_fst_, model_.disambig_);
^~~~~~~~~~~~~~~~~~~
../src/kaldi_recognizer.cc: In constructor ‘KaldiRecognizer::KaldiRecognizer(Model&, float, const char*)’:
../src/kaldi_recognizer.cc:70:9: error: ‘decode_fst_’ was not declared in this scope
decode_fst_ = LookaheadComposeFst(*model_.hcl_fst_, g_fst_, model_.disambig_);
^~~~~~~~~~~
../src/kaldi_recognizer.cc:70:9: note: suggested alternative: ‘decoder_’
decode_fst_ = LookaheadComposeFst(*model_.hcl_fst_, g_fst_, model_.disambig_);
^~~~~~~~~~~
decoder_
../src/kaldi_recognizer.cc:70:23: error: ‘LookaheadComposeFst’ was not declared in this scope
decode_fst_ = LookaheadComposeFst(*model_.hcl_fst_, g_fst_, model_.disambig_);
^~~~~~~~~~~~~~~~~~~
../src/kaldi_recognizer.cc:72:9: error: ‘decode_fst_’ was not declared in this scope
decode_fst_ = NULL;
^~~~~~~~~~~
../src/kaldi_recognizer.cc:72:9: note: suggested alternative: ‘decoder_’
decode_fst_ = NULL;
^~~~~~~~~~~
decoder_
../src/kaldi_recognizer.cc:79:52: error: ‘decode_fst_’ was not declared in this scope
model_.hclg_fst_ ? *model.hclg_fst_ : *decode_fst_,
^~~~~~~~~~~
../src/kaldi_recognizer.cc:79:52: note: suggested alternative: ‘decoder_’
model_.hclg_fst_ ? *model.hclg_fst_ : *decode_fst_,
^~~~~~~~~~~
decoder_
../src/kaldi_recognizer.cc: In destructor ‘KaldiRecognizer::~KaldiRecognizer()’:
../src/kaldi_recognizer.cc:90:12: error: ‘decode_fst_’ was not declared in this scope
delete decode_fst_;
^~~~~~~~~~~
../src/kaldi_recognizer.cc:90:12: note: suggested alternative: ‘decoder_’
delete decode_fst_;
^~~~~~~~~~~
decoder_
../src/kaldi_recognizer.h:50:14: error: ‘LookaheadFst’ in namespace ‘fst’ does not name a template type
fst::LookaheadFst<fst::StdArc, int32> *decode_fst_;
^~~~~~~~~~~~
Makefile:31: recipe for target 'libvosk_jni.so' failed
make: *** [libvosk_jni.so] Error 1
When I compile you app inside a docker container without GPU support everything works fine.
I have made a few changes so that Kaldi is compiled with GPU support and I am running the application inside a docker container with NVIDIA GPU support.
But when I run the GPU version I get the error below, do you have any idea what the problem might be?
server --min-active=200 --max-active=6000 --beam=13.0 --lattice-beam=6.0 --acoustic-scale=1.0 --frame-subsampling-factor=3 --endpoint.silence-phones=1:2:3:4:5:6:7:8:9:10 --endpoint.rule2.min-trailing-silence=0.5 --endpoint.rule3.min-trailing-silence=1.0 --endpoint.rule4.min-trailing-silence=2.0
LOG (server[5.5]:Model():model.cc:47) Sample rate is 8000
LOG (server[5.5]:ComputeDerivedVars():ivector-extractor.cc:183) Computing derived variables for iVector extractor
LOG (server[5.5]:ComputeDerivedVars():ivector-extractor.cc:204) Done.
LOG (server[5.5]:RemoveOrphanNodes():nnet-nnet.cc:948) Removed 1 orphan nodes.
LOG (server[5.5]:RemoveOrphanComponents():nnet-nnet.cc:847) Removing 2 orphan components.
LOG (server[5.5]:Collapse():nnet-utils.cc:1472) Added 1 components, removed 2
LOG (server[5.5]:CompileLooped():nnet-compile-looped.cc:345) Spent 0.0133111 seconds in looped compilation.
WARNING (server[5.5]:SelectGpuId():cu-device.cc:228) Not in compute-exclusive mode. Suggestion: use 'nvidia-smi -c 3' to set compute exclusive mode
LOG (server[5.5]:SelectGpuIdAuto():cu-device.cc:408) Selecting from 1 GPUs
LOG (server[5.5]:SelectGpuIdAuto():cu-device.cc:423) cudaSetDevice(0): GeForce GTX 1070 free:8022M, used:97M, total:8119M, free/total:0.987992
LOG (server[5.5]:SelectGpuIdAuto():cu-device.cc:471) Device: 0, mem_ratio: 0.987992
LOG (server[5.5]:SelectGpuId():cu-device.cc:352) Trying to select device: 0
LOG (server[5.5]:SelectGpuIdAuto():cu-device.cc:481) Success selecting device 0 free mem ratio: 0.987992
LOG (server[5.5]:FinalizeActiveGpu():cu-device.cc:308) The active GPU is [0]: GeForce GTX 1070 free:7834M, used:285M, total:8119M, free/total:0.964838 version 6.1
ERROR (server[5.5]:CopyToMat():cu-matrix.cc:464) cudaError_t 700 : "an illegal memory access was encountered" returned from 'cudaMemcpy2DAsync(dst->Data(), dst_pitch, this->data_, src_pitch, width, this->num_rows_, cudaMemcpyDeviceToHost, cudaStreamPerThread)'
[ Stack-Trace: ]
/opt/kaldi-websocket/_kaldi_recognizer.so(kaldi::MessageLogger::LogMessage() const+0x82c) [0x7f852b86c2de]
/opt/kaldi-websocket/_kaldi_recognizer.so(kaldi::MessageLogger::LogAndThrow::operator=(kaldi::MessageLogger const&)+0x2e) [0x7f852b458f3c]
/opt/kaldi-websocket/_kaldi_recognizer.so(void kaldi::CuMatrixBase<float>::CopyToMat<float>(kaldi::MatrixBase<float>*, kaldi::MatrixTransposeType) const+0x1ea) [0x7f852b78bb52]
/opt/kaldi-websocket/_kaldi_recognizer.so(kaldi::CuMatrix<float>::Swap(kaldi::Matrix<float>*)+0x12f) [0x7f852b78cba5]
/opt/kaldi-websocket/_kaldi_recognizer.so(kaldi::Matrix<float>::Swap(kaldi::CuMatrix<float>*)+0x12) [0x7f852b78cc24]
/opt/kaldi-websocket/_kaldi_recognizer.so(kaldi::nnet3::DecodableNnetLoopedOnlineBase::AdvanceChunk()+0x5a3) [0x7f852b5bbe6f]
/opt/kaldi-websocket/_kaldi_recognizer.so(kaldi::nnet3::DecodableAmNnetLoopedOnline::LogLikelihood(int, int)+0x51) [0x7f852b5bc047]
/opt/kaldi-websocket/_kaldi_recognizer.so(kaldi::LatticeFasterDecoderTpl<fst::VectorFst<fst::ArcTpl<fst::TropicalWeightTpl<float> >, fst::VectorState<fst::ArcTpl<fst::TropicalWeightTpl<float> >, std::allocator<fst::ArcTpl<fst::TropicalWeightTpl<float> > > > >, kaldi::decoder::BackpointerToken>::ProcessEmitting(kaldi::DecodableInterface*)+0x22b) [0x7f852b4c2173]
/opt/kaldi-websocket/_kaldi_recognizer.so(kaldi::LatticeFasterDecoderTpl<fst::VectorFst<fst::ArcTpl<fst::TropicalWeightTpl<float> >, fst::VectorState<fst::ArcTpl<fst::TropicalWeightTpl<float> >, std::allocator<fst::ArcTpl<fst::TropicalWeightTpl<float> > > > >, kaldi::decoder::BackpointerToken>::AdvanceDecoding(kaldi::DecodableInterface*, int)+0x97) [0x7f852b4c2575]
/opt/kaldi-websocket/_kaldi_recognizer.so(kaldi::LatticeFasterDecoderTpl<fst::Fst<fst::ArcTpl<fst::TropicalWeightTpl<float> > >, kaldi::decoder::BackpointerToken>::AdvanceDecoding(kaldi::DecodableInterface*, int)+0x74) [0x7f852b4c30c8]
/opt/kaldi-websocket/_kaldi_recognizer.so(kaldi::SingleUtteranceNnet3DecoderTpl<fst::Fst<fst::ArcTpl<fst::TropicalWeightTpl<float> > > >::AdvanceDecoding()+0x19) [0x7f852b4a1e8f]
/opt/kaldi-websocket/_kaldi_recognizer.so(KaldiRecognizer::AcceptWaveform(char const*, int)+0x10a) [0x7f852b457d8a]
/opt/kaldi-websocket/_kaldi_recognizer.so(+0x2c29c4) [0x7f852b4909c4]
python3(PyCFunction_Call+0x4f) [0x4e12df]
python3(PyEval_EvalFrameEx+0x614) [0x530b94]
python3(PyEval_EvalFrameEx+0x4b64) [0x5350e4]
python3(PyEval_EvalCodeEx+0x13b) [0x53a81b]
python3() [0x4e3537]
python3(PyObject_Call+0x47) [0x5c3bd7]
python3(PyEval_EvalFrameEx+0x24a2) [0x532a22]
python3(PyEval_EvalFrameEx+0x4b64) [0x5350e4]
python3(PyEval_EvalCodeEx+0x13b) [0x53a81b]
python3() [0x4e3537]
python3(PyObject_Call+0x47) [0x5c3bd7]
python3(PyEval_EvalFrameEx+0x24a2) [0x532a22]
python3(PyEval_EvalFrameEx+0x4b64) [0x5350e4]
python3(PyEval_EvalFrameEx+0x4b64) [0x5350e4]
python3(PyEval_EvalCodeEx+0x13b) [0x53a81b]
python3() [0x4e3423]
python3(PyObject_Call+0x47) [0x5c3bd7]
python3() [0x4f08be]
python3(PyObject_Call+0x47) [0x5c3bd7]
python3(PyEval_CallObjectWithKeywords+0x30) [0x525d00]
python3() [0x626bb2]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x76ba) [0x7f852f8816ba]
/lib/x86_64-linux-gnu/libc.so.6(clone+0x6d) [0x7f852f5b741d]
terminate called after throwing an instance of 'kaldi::KaldiFatalError'
Using the fr
model from kaldi-android-demo with vosk 0.3, running the following code crashes the Python interpreter.
from vosk import Model, KaldiRecognizer
try:
model = Model('fr')
rec = KaldiRecognizer(model, 8000)
data = b'\x00' * 1000
print(rec.AcceptWaveform(data))
except Exception as e:
print('Exception!', e)
raise e
print('OK')
root@77cd514bb1ec:/models# python3 /server/bug_repro.py
vosk --min-active=200 --max-active=3000 --beam=10.0 --lattice-beam=2.0 --acoustic-scale=1.0 --frame-subsampling-factor=3 --endpoint.silence-phones=1:2:3:4:5:6:7:8:9:10 --endpoint.rule2.min-trailing-silence=0.5 --endpoint.rule3.min-trailing-silence=1.0 --endpoint.rule4.min-trailing-silence=2.0
LOG (vosk[5.5.599~1-11bed]:ComputeDerivedVars():ivector-extractor.cc:183) Computing derived variables for iVector extractor
LOG (vosk[5.5.599~1-11bed]:ComputeDerivedVars():ivector-extractor.cc:204) Done.
LOG (vosk[5.5.599~1-11bed]:RemoveOrphanNodes():nnet-nnet.cc:948) Removed 1 orphan nodes.
LOG (vosk[5.5.599~1-11bed]:RemoveOrphanComponents():nnet-nnet.cc:847) Removing 2 orphan components.
LOG (vosk[5.5.599~1-11bed]:Collapse():nnet-utils.cc:1472) Added 1 components, removed 2
LOG (vosk[5.5.599~1-11bed]:CompileLooped():nnet-compile-looped.cc:345) Spent 0.0316849 seconds in looped compilation.
ERROR (vosk[5.5.599~1-11bed]:MaybeCreateResampler():online-feature.cc:99) Sampling frequency mismatch, expected 16000, got 8000
Perhaps you want to use the options --allow_{upsample,downsample}
[ Stack-Trace: ]
/usr/local/lib/python3.7/site-packages/vosk/_vosk.so(kaldi::MessageLogger::LogMessage() const+0x7a6) [0x7f1b546ba6ae]
/usr/local/lib/python3.7/site-packages/vosk/_vosk.so(kaldi::MessageLogger::LogAndThrow::operator=(kaldi::MessageLogger const&)+0x13) [0x7f1b5435cd43]
/usr/local/lib/python3.7/site-packages/vosk/_vosk.so(kaldi::OnlineGenericBaseFeature<kaldi::MfccComputer>::MaybeCreateResampler(float)+0x23a) [0x7f1b5456baa2]
/usr/local/lib/python3.7/site-packages/vosk/_vosk.so(kaldi::OnlineGenericBaseFeature<kaldi::MfccComputer>::AcceptWaveform(float, kaldi::VectorBase<float> const&)+0x55) [0x7f1b5456c477]
/usr/local/lib/python3.7/site-packages/vosk/_vosk.so(kaldi::OnlineNnet2FeaturePipeline::AcceptWaveform(float, kaldi::VectorBase<float> const&)+0x1c) [0x7f1b543adc2a]
/usr/local/lib/python3.7/site-packages/vosk/_vosk.so(KaldiRecognizer::AcceptWaveform(kaldi::Vector<float>&)+0x1c) [0x7f1b5435742c]
/usr/local/lib/python3.7/site-packages/vosk/_vosk.so(KaldiRecognizer::AcceptWaveform(char const*, int)+0x17e) [0x7f1b543575ee]
/usr/local/lib/python3.7/site-packages/vosk/_vosk.so(+0x21fe56) [0x7f1b54354e56]
/usr/local/lib/libpython3.7m.so.1.0(_PyMethodDef_RawFastCallKeywords+0x155) [0x7f1b555a64c5]
/usr/local/lib/libpython3.7m.so.1.0(_PyCFunction_FastCallKeywords+0x20) [0x7f1b555a6350]
/usr/local/lib/libpython3.7m.so.1.0(_PyEval_EvalFrameDefault+0x4525) [0x7f1b5561bb95]
/usr/local/lib/libpython3.7m.so.1.0(_PyFunction_FastCallKeywords+0xfa) [0x7f1b555a748a]
/usr/local/lib/libpython3.7m.so.1.0(_PyEval_EvalFrameDefault+0x672) [0x7f1b55617ce2]
/usr/local/lib/libpython3.7m.so.1.0(_PyEval_EvalCodeWithName+0x2f1) [0x7f1b55616b71]
/usr/local/lib/libpython3.7m.so.1.0(PyEval_EvalCodeEx+0x39) [0x7f1b55616879]
/usr/local/lib/libpython3.7m.so.1.0(PyEval_EvalCode+0x1b) [0x7f1b5561683b]
/usr/local/lib/libpython3.7m.so.1.0(+0x20b97e) [0x7f1b5569a97e]
/usr/local/lib/libpython3.7m.so.1.0(PyRun_FileExFlags+0x97) [0x7f1b55699d37]
/usr/local/lib/libpython3.7m.so.1.0(PyRun_SimpleFileExFlags+0x18e) [0x7f1b55699b5e]
/usr/local/lib/libpython3.7m.so.1.0(+0x210108) [0x7f1b5569f108]
/usr/local/lib/libpython3.7m.so.1.0(_Py_UnixMain+0x2e) [0x7f1b5569edfe]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xeb) [0x7f1b5510809b]
python3(_start+0x2a) [0x55f3e267308a]
terminate called after throwing an instance of 'kaldi::KaldiFatalError'
what(): kaldi::KaldiFatalError
Aborted
Hi
I want to know whether this kaldi-android-demo support a model which is not trained with ivector feature. My model works well in this demo, but its size only for the ivector directory is 20M, so I want to test a new model without ivector.
Could you tell me which files are placed insde build/kaldi_${KALDI_SUFFIX}/local and build/kaldi_${KALDI_SUFFIX}/local/include?
... if you will not provide this yourself, maybe you have some hints for possible sources.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.