Is it still possible to install wav2letter (not wav2letter++) in 2021? A lot of links in the wiki are broken and I've been having problems with installation. My end game is to use it for transcriptions of audio files and possible real-time transcription in Python.
Reproduction Steps
Here is a summary of what I've tried:
What I've Tried
This goes temporally, so I don't recall a lot of the earlier errors/problems:
Using Win11
Couldn't get Flashlight, a dependency, to install
Tried compiling binary inference model myself but didn't have all the header files
Using Ubuntu 20.04 LTS with x86_64 ISA on VMware Player
1) Went well until I tried the git remote set-url https://github.com/facebookresearch/wav2letter.git in the "for Inferences pipeline" above the this header, I got a usage error for set-url because two arguments were expected.
2) From here, I tried doing git remote set-url origin https://github.com/facebookresearch/wav2letter.git and moving forward, eventually reaching the error:
Your configuration specifies to merge with the ref 'refs/heads/master'
from the remote, but no such ref was fetched.
3) I tried to build with cmake anyway, which was an apparent success.
4) Then from the build folder, I tried following the Simple Streaming ASR Example from here. From here, I got this error:
cat: /root/audio/LibriSpeech/dev-clean/777/126732/777-126732-0070.flac.wav: No such file or directory
Started features model file loading ...
terminate called after throwing an instance of 'std::runtime_error'
what(): failed to open feature file=/root/model/feature_extractor.bin for reading
Aborted (core dumped)
From here, I shut down and deleted the container
Simple Streaming ASR Example using Docker as laid out here
Here I ran the listed command and received this error:
cat: /root/host/audio/LibriSpeech/dev-clean/777/126732/777-126732-0070.flac.wav: No such file or directory
Started features model file loading ...
terminate called after throwing an instance of 'std::runtime_error'
what(): failed to open feature file=/root/host/model/feature_extractor.bin for reading
Aborted (core dumped)
Here, cloning went fine, but after that I got this error:
unable to prepare context: unable to evaluate symlinks in Dockerfile path: lstat /home/ryan/wav2letter/Dockerfile-Inference: no such file or directory
Then I ran sudo cmake CMakeLists.txt from the wav2letter directory and got this error:
-- The CXX compiler identification is unknown
-- The C compiler identification is GNU 9.3.0
CMake Error at CMakeLists.txt:3 (project):
No CMAKE_CXX_COMPILER could be found.
Tell CMake where to find the compiler by setting either the environment
variable "CXX" or the CMake cache entry CMAKE_CXX_COMPILER to the full path
to the compiler, or to the compiler name if it is in the PATH.
-- Check for working C compiler: /usr/bin/cc
-- Check for working C compiler: /usr/bin/cc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Configuring incomplete, errors occurred!
See also "/home/ryan/wav2letter/CMakeFiles/CMakeOutput.log".
See also "/home/ryan/wav2letter/CMakeFiles/CMakeError.log".
Not using Docker
Download and build wav2letter-inference as seen here
This led to needing MKL and Flashlight. I could not get Flashlight to install. I tried
1) as library with vcpkg
2) from source with vcpkg
3) building from source which lead to problems downloading Flashlight.
benam2
commented
1 year ago
Bug Description
Is it still possible to install wav2letter (not wav2letter++) in 2021? A lot of links in the wiki are broken and I've been having problems with installation. My end game is to use it for transcriptions of audio files and possible real-time transcription in Python.
Reproduction Steps
Here is a summary of what I've tried:
What I've Tried
This goes temporally, so I don't recall a lot of the earlier errors/problems:
Using Win11
Using Ubuntu 20.04 LTS with x86_64 ISA on VMware Player
Using pre-built Docker images as laid out here
1) Went well until I tried the
git remote set-url https://github.com/facebookresearch/wav2letter.gitin the "for Inferences pipeline" above the this header, I got a usage error forset-urlbecause two arguments were expected.2) From here, I tried doing
git remote set-url origin https://github.com/facebookresearch/wav2letter.gitand moving forward, eventually reaching the error:3) I tried to build with cmake anyway, which was an apparent success. 4) Then from the
buildfolder, I tried following the Simple Streaming ASR Example from here. From here, I got this error:From here, I shut down and deleted the container
Simple Streaming ASR Example using Docker as laid out here
Here I ran the listed command and received this error:
Building the Docker image myself as seen here
Here, cloning went fine, but after that I got this error:
Then I ran
sudo cmake CMakeLists.txtfrom thewav2letterdirectory and got this error:Not using Docker
Download and build wav2letter-inference as seen here
This led to needing MKL and Flashlight. I could not get Flashlight to install. I tried 1) as library with vcpkg 2) from source with vcpkg 3) building from source which lead to problems downloading Flashlight.
Lastly, I tried following this tutorial
Eventually running into an error, I believe installing Flashlight
Platform and Hardware
Also attempted installation on a VM
Additional Context
[Add any additional information here]