Wav2letter: Python bindings

Created on 9 Sep 2020 · 22Comments · Source: flashlight/wav2letter

I got an error message below when I try ~/wav2letter/bindings/python$ pip install -e .


  Running setup.py develop for wav2letter
    ERROR: Command errored out with exit status 1:
     command: /opt/conda/bin/python -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/root/wa                                                                                                                        v2letter/bindings/python/setup.py'"'"'; __file__='"'"'/root/wav2letter/bindings/python/setup.py'"'"';                                                                                                                        f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"                                                                                                                        ');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' develop --no-deps
         cwd: /root/wav2letter/bindings/python/
    Complete output (36 lines):
    running develop
    running egg_info
    writing wav2letter.egg-info/PKG-INFO
    writing dependency_links to wav2letter.egg-info/dependency_links.txt
    writing top-level names to wav2letter.egg-info/top_level.txt
    reading manifest file 'wav2letter.egg-info/SOURCES.txt'
    writing manifest file 'wav2letter.egg-info/SOURCES.txt'
    running build_ext
    Error: could not load cache
    Traceback (most recent call last):
      File "<string>", line 1, in <module>
      File "/root/wav2letter/bindings/python/setup.py", line 109, in <module>
        zip_safe=False,
      File "/opt/conda/lib/python3.7/site-packages/setuptools/__init__.py", line 144, in setup
        return distutils.core.setup(**attrs)
      File "/opt/conda/lib/python3.7/distutils/core.py", line 148, in setup
        dist.run_commands()
      File "/opt/conda/lib/python3.7/distutils/dist.py", line 966, in run_commands
        self.run_command(cmd)
      File "/opt/conda/lib/python3.7/distutils/dist.py", line 985, in run_command
        cmd_obj.run()
      File "/opt/conda/lib/python3.7/site-packages/setuptools/command/develop.py", line 38, in run
        self.install_for_development()
      File "/opt/conda/lib/python3.7/site-packages/setuptools/command/develop.py", line 140, in insta                                                                                                                        ll_for_development
        self.run_command('build_ext')
      File "/opt/conda/lib/python3.7/distutils/cmd.py", line 313, in run_command
        self.distribution.run_command(command)
      File "/opt/conda/lib/python3.7/distutils/dist.py", line 985, in run_command
        cmd_obj.run()
      File "/root/wav2letter/bindings/python/setup.py", line 48, in run
        self.build_extensions()
      File "/root/wav2letter/bindings/python/setup.py", line 91, in build_extensions
        ["cmake", "--build", "."] + build_args, cwd=self.build_temp
      File "/opt/conda/lib/python3.7/subprocess.py", line 363, in check_call
        raise CalledProcessError(retcode, cmd)
    subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--config', 'Release', '--', '-                                                                                                                        j4']' returned non-zero exit status 1.
    ----------------------------------------
ERROR: Command errored out with exit status 1: /opt/conda/bin/python -c 'import sys, setuptools, toke                                                                                                                        nize; sys.argv[0] = '"'"'/root/wav2letter/bindings/python/setup.py'"'"'; __file__='"'"'/root/wav2lett                                                                                                                        er/bindings/python/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().re                                                                                                                        place('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' develop                                                                                                                         --no-deps Check the logs for full command output.

How can I bind python? Plz help me.

question

Source

sooftware

All 22 comments

Oh, I solved it. Never mind.
I solved it by installed with the previous commit.

sooftware on 9 Sep 2020

👍1

Oh, I solved it. Never mind.
I solved it by installed with the previous commit.

Which commit, can you give the ID?

HJJ256 on 10 Sep 2020

Please use this branch latest commit https://github.com/facebookresearch/wav2letter/tree/v0.2

tlikhomanenko on 11 Sep 2020

👍1

I have got the same error, could you please share how you managed to solve it

abkw on 2 Feb 2021

Has someone able to solve this error? I am not able to solve this error! I tried the suggested solution of removing the already existing build folder but the same error is coming up again. @tlikhomanenko : could you help?

gauravgund on 22 Mar 2021

Could you specify what fl / w2l versions are you using?

tlikhomanenko on 22 Mar 2021

w2l: v.0.2
I am referring these instructions:
[https://github.com/mailong25/self-supervised-speech-recognition/blob/master/Dependencies.md]

gauravgund on 22 Mar 2021

Could you post here your full error log on pip install -e .?

tlikhomanenko on 22 Mar 2021

I am running it in google colab.
The code is as follows:

%%bash git clone -b v0.2 https://github.com/facebookresearch/wav2letter.git cd wav2letter/bindings/python export KENLM_ROOT_DIR=/content/kenlm && pip install -e . cd ../../..

The error is as follows:
_ERROR: Command errored out with exit status 1: /usr/bin/python3 -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/wav2letter/bindings/python/setup.py'"'"'; __file__='"'"'/wav2letter/bindings/python/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' develop --no-deps Check the logs for full command output.

gauravgund on 23 Mar 2021

I would recommend to switch to the latest fl version and its python bindings. But if it is crucial for you to use v0.2 can you send simple colab where I can repro the error so that I can help on its resolving?

tlikhomanenko on 27 Mar 2021

https://github.com/gauravgund/gauravgund/blob/master/wav2letter.ipynb

Here is the colab notebook!
Thanks.

gauravgund on 27 Mar 2021

Ok, you have several problems:
1) for some reason log of pip is not printed into colab cell so you don't see actual log and errors. To fix this you can add pip install --log=pip_log -e . - log will be in the folder from where you run pip install and you can see actual error
2) You need to fix kenlm path to the actual path you have for it: export KENLM_ROOT_DIR=/content/self-supervised-speech-recognition/libs/kenlm
3) with cuda criterion compilation (part of bindings) there is error on unsupported arch: nvcc fatal : Unsupported gpu architecture 'compute_30' so, either use another gpus or if you don't need ASG criterion from bindings simply set export USE_CUDA=OFF

To sum up:

%%bash
rm -rf wav2letter
git clone -b v0.2 https://github.com/facebookresearch/wav2letter.git
cd wav2letter/bindings/python
export KENLM_ROOT_DIR=/content/self-supervised-speech-recognition/libs/kenlm
export USE_CUDA=0
pip install --log pip_log -e .
cd ../../..

This works for me and I see in the cell

Obtaining file:///content/self-supervised-speech-recognition/libs/wav2letter/bindings/python
Installing collected packages: wav2letter
  Running setup.py develop for wav2letter
Successfully installed wav2letter

Cloning into 'wav2letter'...

and imports are working for me.

tlikhomanenko on 28 Mar 2021

Thanks a lot for your help! Let me try it again and I will get back to you!

Thanks again! :)

Regards

On Sun, 28 Mar 2021 at 12:34 PM, Tatiana Likhomanenko <
@.*> wrote:

Ok, you have several problems:

for some reason log of pip is not printed into colab cell so you
don't see actual log and errors. To fix this you can add pip install
--log=pip_log -e . - log will be in the folder from where you run pip
install and you can see actual error

You need to fix kenlm path to the actual path you have for it: export
KENLM_ROOT_DIR=/content/self-supervised-speech-recognition/libs/kenlm

with cuda criterion compilation (part of bindings) there is error
on unsupported arch: nvcc fatal : Unsupported gpu architecture
'compute_30' so, either use another gpus or if you don't need ASG
criterion from bindings simply set export USE_CUDA=OFF

To sum up:

%%bash
rm -rf wav2letter
git clone -b v0.2 https://github.com/facebookresearch/wav2letter.git
cd wav2letter/bindings/python
export KENLM_ROOT_DIR=/content/self-supervised-speech-recognition/libs/kenlm
export USE_CUDA=0
pip install --log pip_log -e .
cd ../../..

This works for me and I see in the cell

Obtaining file:///content/self-supervised-speech-recognition/libs/wav2letter/bindings/python
Installing collected packages: wav2letter
Running setup.py develop for wav2letter
Successfully installed wav2letter

Cloning into 'wav2letter'...

and imports are working for me.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
https://github.com/facebookresearch/wav2letter/issues/819#issuecomment-808857563,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/ALBKX6UWIHME5KJ35SFQHBLTF3IJNANCNFSM4RCZ54QA
.

gauravgund on 28 Mar 2021

HI @tlikhomanenko : I am still getting the same error. As recommended i fixed the Cuda and export kenlm correctly. But still the error is the same. Also, I am running pip install --log in the 'wav2letter/bindings/python' directory but i could not find any file by the name pip_log.

Code:
%%bash
rm -rf wav2letter
git clone -b v0.2 https://github.com/facebookresearch/wav2letter.git
cd wav2letter/bindings/python
export KENLM_ROOT_DIR=/content/self-supervised-speech-recognition/libs/kenlm
pip install --log pip_log -e .
cd ../../..

Error:
Obtaining file:///content/self-supervised-speech-recognition/libs/wav2letter/bindings/python
Installing collected packages: wav2letter
Running setup.py develop for wav2letter
Cloning into 'wav2letter'...
ERROR: Command errored out with exit status 1: /usr/bin/python3 -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/setup.py'"'"'; __file__='"'"'/content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' develop --no-deps Check the logs for full command output.

gauravgund on 29 Mar 2021

pip log will be in the wav2letter/bindings/python, also please first try with USE_CUDA=OFF.

Could you send ls -lah on wav2letter/bindings/python?

tlikhomanenko on 30 Mar 2021

HI @tlikhomanenko: It worked with export USE_CUDA=0. How to set this up with cuda now?

gauravgund on 30 Mar 2021

Here is the log when USE_CUDA=1 is used:

2021-03-30T17:21:54,670 Created temporary directory: /tmp/pip-ephem-wheel-cache-m1yywrax
2021-03-30T17:21:54,671 Created temporary directory: /tmp/pip-req-tracker-_6fe_m5x
2021-03-30T17:21:54,671 Created requirements tracker '/tmp/pip-req-tracker-_6fe_m5x'
2021-03-30T17:21:54,671 Created temporary directory: /tmp/pip-install-bk2swb_0
2021-03-30T17:21:54,672 Obtaining file:///content/self-supervised-speech-recognition/libs/wav2letter/bindings/python
2021-03-30T17:21:54,672 Added file:///content/self-supervised-speech-recognition/libs/wav2letter/bindings/python to build tracker '/tmp/pip-req-tracker-_6fe_m5x'
2021-03-30T17:21:54,672 Running setup.py (path:/content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/setup.py) egg_info for package from file:///content/self-supervised-speech-recognition/libs/wav2letter/bindings/python
2021-03-30T17:21:54,672 Running command python setup.py egg_info
2021-03-30T17:21:55,000 running egg_info
2021-03-30T17:21:55,000 creating wav2letter.egg-info
2021-03-30T17:21:55,001 writing wav2letter.egg-info/PKG-INFO
2021-03-30T17:21:55,001 writing dependency_links to wav2letter.egg-info/dependency_links.txt
2021-03-30T17:21:55,001 writing top-level names to wav2letter.egg-info/top_level.txt
2021-03-30T17:21:55,001 writing manifest file 'wav2letter.egg-info/SOURCES.txt'
2021-03-30T17:21:55,017 writing manifest file 'wav2letter.egg-info/SOURCES.txt'
2021-03-30T17:21:55,040 Source in /content/self-supervised-speech-recognition/libs/wav2letter/bindings/python has version 0.0.2, which satisfies requirement wav2letter==0.0.2 from file:///content/self-supervised-speech-recognition/libs/wav2letter/bindings/python
2021-03-30T17:21:55,040 Removed wav2letter==0.0.2 from file:///content/self-supervised-speech-recognition/libs/wav2letter/bindings/python from build tracker '/tmp/pip-req-tracker-_6fe_m5x'
2021-03-30T17:21:56,586 Installing collected packages: wav2letter
2021-03-30T17:21:56,586 Running setup.py develop for wav2letter
2021-03-30T17:21:56,586 Running command /usr/bin/python3 -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/setup.py'"'"'; __file__='"'"'/content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' develop --no-deps
2021-03-30T17:21:56,925 running develop
2021-03-30T17:21:57,094 running egg_info
2021-03-30T17:21:57,094 writing wav2letter.egg-info/PKG-INFO
2021-03-30T17:21:57,095 writing dependency_links to wav2letter.egg-info/dependency_links.txt
2021-03-30T17:21:57,095 writing top-level names to wav2letter.egg-info/top_level.txt
2021-03-30T17:21:57,111 writing manifest file 'wav2letter.egg-info/SOURCES.txt'
2021-03-30T17:21:57,111 running build_ext
2021-03-30T17:21:57,254 -- The C compiler identification is GNU 7.5.0
2021-03-30T17:21:57,331 -- The CXX compiler identification is GNU 7.5.0
2021-03-30T17:21:57,341 -- Check for working C compiler: /usr/bin/cc
2021-03-30T17:21:57,419 -- Check for working C compiler: /usr/bin/cc -- works
2021-03-30T17:21:57,420 -- Detecting C compiler ABI info
2021-03-30T17:21:57,499 -- Detecting C compiler ABI info - done
2021-03-30T17:21:57,515 -- Detecting C compile features
2021-03-30T17:21:57,753 -- Detecting C compile features - done
2021-03-30T17:21:57,760 -- Check for working CXX compiler: /usr/bin/c++
2021-03-30T17:21:57,848 -- Check for working CXX compiler: /usr/bin/c++ -- works
2021-03-30T17:21:57,849 -- Detecting CXX compiler ABI info
2021-03-30T17:21:57,938 -- Detecting CXX compiler ABI info - done
2021-03-30T17:21:57,953 -- Detecting CXX compile features
2021-03-30T17:21:58,363 -- Detecting CXX compile features - done
2021-03-30T17:21:58,461 -- Looking for pthread.h
2021-03-30T17:21:58,542 -- Looking for pthread.h - found
2021-03-30T17:21:58,543 -- Looking for pthread_create
2021-03-30T17:21:58,624 -- Looking for pthread_create - not found
2021-03-30T17:21:58,624 -- Looking for pthread_create in pthreads
2021-03-30T17:21:58,691 -- Looking for pthread_create in pthreads - not found
2021-03-30T17:21:58,691 -- Looking for pthread_create in pthread
2021-03-30T17:21:58,769 -- Looking for pthread_create in pthread - found
2021-03-30T17:21:58,772 -- Found Threads: TRUE
2021-03-30T17:21:58,796 -- CUDA found (library: /usr/local/cuda/lib64/libcudart_static.a;-lpthread;dl;/usr/lib/x86_64-linux-gnu/librt.so include: /usr/local/cuda/include)
2021-03-30T17:21:58,800 -- CUDA architecture flags: -gencodearch=compute_30,code=sm_30-gencodearch=compute_35,code=sm_35-gencodearch=compute_50,code=sm_50-gencodearch=compute_52,code=sm_52-gencodearch=compute_60,code=sm_60-gencodearch=compute_61,code=sm_61-gencodearch=compute_70,code=sm_70-gencodearch=compute_75,code=sm_75-gencodearch=compute_70,code=compute_70-gencodearch=compute_75,code=compute_75
2021-03-30T17:21:58,829 -- Found PkgConfig: /usr/bin/pkg-config (found version "0.29.1")
2021-03-30T17:21:58,831 -- Checking for module 'cblas'
2021-03-30T17:21:58,843 -- No package 'cblas' found
2021-03-30T17:21:58,845 -- Looking for sys/types.h
2021-03-30T17:21:58,925 -- Looking for sys/types.h - found
2021-03-30T17:21:58,926 -- Looking for stdint.h
2021-03-30T17:21:59,004 -- Looking for stdint.h - found
2021-03-30T17:21:59,004 -- Looking for stddef.h
2021-03-30T17:21:59,081 -- Looking for stddef.h - found
2021-03-30T17:21:59,081 -- Check size of void*
2021-03-30T17:21:59,163 -- Check size of void* - done
2021-03-30T17:21:59,166 -- Checking for [Accelerate]
2021-03-30T17:21:59,172 -- Checking for [vecLib]
2021-03-30T17:21:59,175 -- Checking for [cblas - atlas]
2021-03-30T17:21:59,178 -- Includes found
2021-03-30T17:21:59,179 -- Looking for cblas_dgemm
2021-03-30T17:21:59,268 -- Looking for cblas_dgemm - found
2021-03-30T17:21:59,268 -- CBLAS Symbols FOUND
2021-03-30T17:21:59,269 -- CBLAS library found
2021-03-30T17:21:59,269 -- CBLAS found (include: /usr/include/x86_64-linux-gnu, library: /usr/lib/x86_64-linux-gnu/libcblas.so;/usr/lib/x86_64-linux-gnu/libatlas.so)
2021-03-30T17:21:59,357 -- Found FFTW: /usr/include
2021-03-30T17:21:59,357 -- FFTW found
2021-03-30T17:21:59,358 -- Looking for KenLM
2021-03-30T17:21:59,361 -- Using kenlm library found in /content/self-supervised-speech-recognition/libs/kenlm/build/lib/libkenlm.a
2021-03-30T17:21:59,361 -- Using kenlm utils library found in /content/self-supervised-speech-recognition/libs/kenlm/build/lib/libkenlm.a
2021-03-30T17:21:59,361 -- kenlm lm/model.hh found in /content/self-supervised-speech-recognition/libs/kenlm/lm/model.hh
2021-03-30T17:21:59,363 -- Found kenlm: /content/self-supervised-speech-recognition/libs/kenlm
2021-03-30T17:21:59,363 -- Found kenlm (include: /content/self-supervised-speech-recognition/libs/kenlm, library: /content/self-supervised-speech-recognition/libs/kenlm/build/lib/libkenlm.a;/content/self-supervised-speech-recognition/libs/kenlm/build/lib/libkenlm_util.a)
2021-03-30T17:21:59,365 -- Looking for lzma_auto_decoder in /usr/lib/x86_64-linux-gnu/liblzma.so
2021-03-30T17:21:59,445 -- Looking for lzma_auto_decoder in /usr/lib/x86_64-linux-gnu/liblzma.so - found
2021-03-30T17:21:59,446 -- Looking for lzma_easy_encoder in /usr/lib/x86_64-linux-gnu/liblzma.so
2021-03-30T17:21:59,529 -- Looking for lzma_easy_encoder in /usr/lib/x86_64-linux-gnu/liblzma.so - found
2021-03-30T17:21:59,529 -- Looking for lzma_lzma_preset in /usr/lib/x86_64-linux-gnu/liblzma.so
2021-03-30T17:21:59,612 -- Looking for lzma_lzma_preset in /usr/lib/x86_64-linux-gnu/liblzma.so - found
2021-03-30T17:21:59,615 -- Found LibLZMA: /usr/include (found version "5.2.2")
2021-03-30T17:21:59,622 -- Found BZip2: /usr/lib/x86_64-linux-gnu/libbz2.so (found version "1.0.6")
2021-03-30T17:21:59,623 -- Looking for BZ2_bzCompressInit
2021-03-30T17:21:59,705 -- Looking for BZ2_bzCompressInit - found
2021-03-30T17:21:59,718 -- Found ZLIB: /usr/lib/x86_64-linux-gnu/libz.so (found version "1.2.11")
2021-03-30T17:21:59,838 -- Found PythonInterp: /usr/bin/python3 (found version "3.7.10")
2021-03-30T17:21:59,878 -- Found PythonLibs: /usr/lib/x86_64-linux-gnu/libpython3.7m.so
2021-03-30T17:21:59,892 -- Performing Test HAS_FLTO
2021-03-30T17:22:00,213 -- Performing Test HAS_FLTO - Success
2021-03-30T17:22:00,214 -- LTO enabled
2021-03-30T17:22:00,221 -- Configuring done
2021-03-30T17:22:00,328 -- Generating done
2021-03-30T17:22:00,333 -- Build files have been written to: /content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/build/temp.linux-x86_64-3.7
2021-03-30T17:22:00,400 Scanning dependencies of target CUB
2021-03-30T17:22:00,401 Scanning dependencies of target pybind11
2021-03-30T17:22:00,410 [ 0%] Creating directories for 'CUB'
2021-03-30T17:22:00,410 [ 1%] Creating directories for 'pybind11'
2021-03-30T17:22:00,454 [ 1%] Performing download step (git clone) for 'pybind11'
2021-03-30T17:22:00,455 [ 2%] Performing download step (git clone) for 'CUB'
2021-03-30T17:22:00,467 Cloning into 'pybind11'...
2021-03-30T17:22:00,467 Cloning into 'CUB'...
2021-03-30T17:22:01,570 Note: checking out '9a19306fbf30642ca331d0ec88e7da54a96860f9'.

2021-03-30T17:22:01,570 You are in 'detached HEAD' state. You can look around, make experimental
2021-03-30T17:22:01,570 changes and commit them, and you can discard any commits you make in this
2021-03-30T17:22:01,570 state without impacting any branches by performing another checkout.

2021-03-30T17:22:01,570 If you want to create a new branch to retain commits you create, you may
2021-03-30T17:22:01,570 do so (now or later) by using -b with the checkout command again. Example:

2021-03-30T17:22:01,570 git checkout -b

2021-03-30T17:22:01,570 HEAD is now at 9a19306 bump version to 2.2.4
2021-03-30T17:22:01,640 Submodule 'tools/clang' (https://github.com/wjakob/clang-cindex-python3) registered for path 'tools/clang'
2021-03-30T17:22:01,703 Cloning into '/content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/build/temp.linux-x86_64-3.7/bindings/python/pybind11/src/pybind11/tools/clang'...
2021-03-30T17:22:02,370 Submodule path 'tools/clang': checked out '6a00cbc4a9b8e68b71caf7f774b3f9c753ae84d5'
2021-03-30T17:22:02,409 [ 2%] No patch step for 'pybind11'
2021-03-30T17:22:02,416 [ 3%] Performing update step for 'pybind11'
2021-03-30T17:22:02,460 [ 4%] No configure step for 'pybind11'
2021-03-30T17:22:02,482 [ 5%] No build step for 'pybind11'
2021-03-30T17:22:02,502 [ 6%] No install step for 'pybind11'
2021-03-30T17:22:02,522 [ 6%] Completed 'pybind11'
2021-03-30T17:22:02,549 [ 6%] Built target pybind11
2021-03-30T17:22:03,245 Note: checking out 'c3cceac115c072fb63df1836ff46d8c60d9eb304'.

2021-03-30T17:22:03,245 You are in 'detached HEAD' state. You can look around, make experimental
2021-03-30T17:22:03,245 changes and commit them, and you can discard any commits you make in this
2021-03-30T17:22:03,245 state without impacting any branches by performing another checkout.

2021-03-30T17:22:03,245 If you want to create a new branch to retain commits you create, you may
2021-03-30T17:22:03,245 do so (now or later) by using -b with the checkout command again. Example:

2021-03-30T17:22:03,245 git checkout -b

2021-03-30T17:22:03,245 HEAD is now at c3cceac1 update readme to 1.8.0
2021-03-30T17:22:03,358 [ 7%] No patch step for 'CUB'
2021-03-30T17:22:03,358 [ 7%] Performing update step for 'CUB'
2021-03-30T17:22:03,381 [ 8%] No configure step for 'CUB'
2021-03-30T17:22:03,396 [ 8%] No build step for 'CUB'
2021-03-30T17:22:03,411 [ 9%] No install step for 'CUB'
2021-03-30T17:22:03,425 [ 9%] Completed 'CUB'
2021-03-30T17:22:03,446 [ 9%] Built target CUB
2021-03-30T17:22:03,455 [ 10%] Building NVCC (Device) object src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/w2l-criterion-library-cuda_generated_CriterionUtils.cu.o
2021-03-30T17:22:03,456 [ 10%] Building NVCC (Device) object src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/w2l-criterion-library-cuda_generated_ViterbiPath.cu.o
2021-03-30T17:22:03,456 [ 11%] Building NVCC (Device) object src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/w2l-criterion-library-cuda_generated_ForceAlignmentCriterion.cu.o
2021-03-30T17:22:03,457 [ 11%] Building NVCC (Device) object src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/w2l-criterion-library-cuda_generated_FullConnectionCriterion.cu.o
2021-03-30T17:22:03,480 nvcc fatal : Unsupported gpu architecture 'compute_30'
2021-03-30T17:22:03,480 nvcc fatal : Unsupported gpu architecture 'compute_30'
2021-03-30T17:22:03,482 CMake Error at w2l-criterion-library-cuda_generated_ViterbiPath.cu.o.Release.cmake:219 (message):
2021-03-30T17:22:03,482 Error generating
2021-03-30T17:22:03,482 /content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/build/temp.linux-x86_64-3.7/src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/./w2l-criterion-library-cuda_generated_ViterbiPath.cu.o

2021-03-30T17:22:03,482 CMake Error at w2l-criterion-library-cuda_generated_CriterionUtils.cu.o.Release.cmake:219 (message):
2021-03-30T17:22:03,482 Error generating
2021-03-30T17:22:03,482 /content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/build/temp.linux-x86_64-3.7/src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/./w2l-criterion-library-cuda_generated_CriterionUtils.cu.o

2021-03-30T17:22:03,482 nvcc fatal : Unsupported gpu architecture 'compute_30'
2021-03-30T17:22:03,483 src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/build.make:84: recipe for target 'src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/w2l-criterion-library-cuda_generated_ViterbiPath.cu.o' failed
2021-03-30T17:22:03,483 make[2]: * [src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/w2l-criterion-library-cuda_generated_ViterbiPath.cu.o] Error 1
2021-03-30T17:22:03,483 make[2]: Waiting for unfinished jobs....
2021-03-30T17:22:03,483 src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/build.make:63: recipe for target 'src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/w2l-criterion-library-cuda_generated_CriterionUtils.cu.o' failed
2021-03-30T17:22:03,483 make[2]: ** [src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/w2l-criterion-library-cuda_generated_CriterionUtils.cu.o] Error 1
2021-03-30T17:22:03,483 CMake Error at w2l-criterion-library-cuda_generated_ForceAlignmentCriterion.cu.o.Release.cmake:219 (message):
2021-03-30T17:22:03,483 Error generating
2021-03-30T17:22:03,483 /content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/build/temp.linux-x86_64-3.7/src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/./w2l-criterion-library-cuda_generated_ForceAlignmentCriterion.cu.o

2021-03-30T17:22:03,483 nvcc fatal : Unsupported gpu architecture 'compute_30'
2021-03-30T17:22:03,483 CMake Error at w2l-criterion-library-cuda_generated_FullConnectionCriterion.cu.o.Release.cmake:219 (message):
2021-03-30T17:22:03,483 Error generating
2021-03-30T17:22:03,483 /content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/build/temp.linux-x86_64-3.7/src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/./w2l-criterion-library-cuda_generated_FullConnectionCriterion.cu.o

2021-03-30T17:22:03,484 2021-03-30T17:22:03,484 2021-03-30T17:22:03,484 2021-03-30T17:22:03,484 2021-03-30T17:22:03,484 2021-03-30T17:22:03,485 2021-03-30T17:22:03,485 2021-03-30T17:22:03,485 2021-03-30T17:22:03,495 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,496 2021-03-30T17:22:03,497 2021-03-30T17:22:03,497 2021-03-30T17:22:03,497 2021-03-30T17:22:03,497 2021-03-30T17:22:03,497 2021-03-30T17:22:03,497 2021-03-30T17:22:03,497 2021-03-30T17:22:03,497 2021-03-30T17:22:03,526 2021-03-30T17:22:03,526 2021-03-30T17:22:03,526 2021-03-30T17:22:03,527 2021-03-30T17:22:03,527 2021-03-30T17:22:03,527 2021-03-30T17:22:03,527 2021-03-30T17:22:03,527 2021-03-30T17:22:03,527 2021-03-30T17:22:03,527 2021-03-30T17:22:03,527 2021-03-30T17:22:03,527 2021-03-30T17:22:03,527 2021-03-30T17:22:03,527 2021-03-30T17:22:03,527 2021-03-30T17:22:03,527 2021-03-30T17:22:03,527 2021-03-30T17:22:03,527 src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/build.make:70: recipe for target 'src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/w2l-criterion-library-cuda_generated_ForceAlignmentCriterion.cu.o' failed
make[2]: * [src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/w2l-criterion-library-cuda_generated_ForceAlignmentCriterion.cu.o] Error 1
src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/build.make:77: recipe for target 'src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/w2l-criterion-library-cuda_generated_FullConnectionCriterion.cu.o' failed
make[2]: [src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/cuda/w2l-criterion-library-cuda_generated_FullConnectionCriterion.cu.o] Error 1
CMakeFiles/Makefile2:179: recipe for target 'src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/all' failed
make[1]: [src/libraries/criterion/CMakeFiles/w2l-criterion-library-cuda.dir/all] Error 2
Makefile:83: recipe for target 'all' failed
make: [all] Error 2
Traceback (most recent call last):
File "", line 1, in
File "/content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/setup.py", line 109, in
zip_safe=False,
File "/usr/local/lib/python3.7/dist-packages/setuptools/__init__.py", line 153, in setup
return distutils.core.setup(attrs)
File "/usr/lib/python3.7/distutils/core.py", line 148, in setup
dist.run_commands()
File "/usr/lib/python3.7/distutils/dist.py", line 966, in run_commands
self.run_command(cmd)
File "/usr/lib/python3.7/distutils/dist.py", line 985, in run_command
cmd_obj.run()
File "/usr/local/lib/python3.7/dist-packages/setuptools/command/develop.py", line 34, in run
self.install_for_development()
File "/usr/local/lib/python3.7/dist-packages/setuptools/command/develop.py", line 136, in install_for_development
self.run_command('build_ext')
File "/usr/lib/python3.7/distutils/cmd.py", line 313, in run_command
self.distribution.run_command(command)
File "/usr/lib/python3.7/distutils/dist.py", line 985, in run_command
cmd_obj.run()
File "/content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/setup.py", line 48, in run
self.build_extensions()
File "/content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/setup.py", line 91, in build_extensions
["cmake", "--build", "."] + build_args, cwd=self.build_temp
File "/usr/lib/python3.7/subprocess.py", line 363, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--config', 'Release', '--', '-j4']' returned non-zero exit status 2.
Cleaning up...
Removed build tracker '/tmp/pip-req-tracker-_6fe_m5x'
ERROR: Command errored out with exit status 1: /usr/bin/python3 -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/setup.py'"'"'; __file__='"'"'/content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' develop --no-deps Check the logs for full command output.
Exception information:
Traceback (most recent call last):
File "/usr/local/lib/python3.7/dist-packages/pip/_internal/cli/base_command.py", line 153, in _main
status = self.run(options, args)
File "/usr/local/lib/python3.7/dist-packages/pip/_internal/commands/install.py", line 455, in run
use_user_site=options.use_user_site,
File "/usr/local/lib/python3.7/dist-packages/pip/_internal/req/__init__.py", line 62, in install_given_reqs
*kwargs
File "/usr/local/lib/python3.7/dist-packages/pip/_internal/req/req_install.py", line 851, in install
install_options, global_options, prefix=prefix,
File "/usr/local/lib/python3.7/dist-packages/pip/_internal/req/req_install.py", line 700, in install_editable
cwd=self.unpacked_source_directory,
File "/usr/local/lib/python3.7/dist-packages/pip/_internal/utils/subprocess.py", line 242, in call_subprocess
raise InstallationError(exc_msg)
pip._internal.exceptions.InstallationError: Command errored out with exit status 1: /usr/bin/python3 -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/setup.py'"'"'; __file__='"'"'/content/self-supervised-speech-recognition/libs/wav2letter/bindings/python/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' develop --no-deps Check the logs for full command output.

gauravgund on 30 Mar 2021

the problem with the underlying cuda arch nvcc fatal : Unsupported gpu architecture 'compute_30'.

can you check what gpu do you have there?
do you need ASG criterion or why do you need python bindings with CUDA?

tlikhomanenko on 1 Apr 2021

I am able to resolve that error by downgrading to cuda 10.0.

I have trained the model in pytorch format but i am not able to find a way
to transcribe .wav files with the custom language model. Any inputs on
that would be appreciated. :) Thanks a lot for your help!

On Thu, Apr 1, 2021 at 5:08 AM Tatiana Likhomanenko <
@.*> wrote:

the problem with the underlying cuda arch nvcc fatal : Unsupported gpu
architecture 'compute_30'.

can you check what gpu do you have there?

do you need ASG criterion or why do you need python bindings with
CUDA?

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
https://github.com/facebookresearch/wav2letter/issues/819#issuecomment-811533287,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/ALBKX6R4LQGTXAECFVCYCVLTGOW6TANCNFSM4RCZ54QA
.

gauravgund on 1 Apr 2021

Please read docs on python bindings and here is an example how to define your own class https://github.com/facebookresearch/flashlight/tree/master/bindings/python#define-your-own-language-model-for-beam-search-decoding with custom lm. But first I would suggest to try decoding with LM trained in kenlm library. All doc with examples is here https://github.com/facebookresearch/flashlight/tree/master/bindings/python.

tlikhomanenko on 5 Apr 2021

Hi,

I am looking for light-weight acoustic models to do fine-tuning! The
transformers model with 70 million was too heavy for 1 GPU. Could you throw
some light in this direction?

Regards

On Tue, 6 Apr 2021 at 12:00 AM, Tatiana Likhomanenko <
@.*> wrote:

Please read docs on python bindings and here is an example how to define
your own class
https://github.com/facebookresearch/flashlight/tree/master/bindings/python#define-your-own-language-model-for-beam-search-decoding
with custom lm. But first I would suggest to try decoding with LM trained
in kenlm library. All doc with examples is here
https://github.com/facebookresearch/flashlight/tree/master/bindings/python
.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
https://github.com/facebookresearch/wav2letter/issues/819#issuecomment-813561205,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/ALBKX6QVMO4LPGGKFOFC56LTHH6TJANCNFSM4RCZ54QA
.