shoegazerstella/deepspeech_cpu_py37

A CPU version of DeepSpeech for python 3.7 🗣️📝

DeepSpeech CPU on python 3.7

This version of PaddlePaddle DeepSpeech is intended for inference only to run on CPU with python 3.7

Docker usage

docker build -f Dockerfile_cpu_py37 -t deepspeech-cpu .
docker run -v /your/data/and/model/dir/:/root/DeepSpeech/DATA/ -it deepspeech-cpu

make sure to use the flag --use_gpu=False when calling infer.py

deepspeech_cpu_py37's Issues

Docker building error.

openfst-1.6.3/src/include/fst/string.h:107:55: note: declared here
enum OPENFST_DEPRECATED("Use fst::StringTokenType") TokenType {
^~~~~~~~~
openfst-1.6.3/src/include/fst/string.h:207:59: warning: ‘TokenType’ is deprecated: Use fst::StringTokenType [-Wdeprecated-declarations]
const SymbolTable *syms = nullptr)
^
openfst-1.6.3/src/include/fst/string.h:199:55: note: declared here
enum OPENFST_DEPRECATED("Use fst::StringTokenType") TokenType {
^~~~~~~~~
error: command 'gcc' failed with exit status 1

Any help for this issue?

[Feature Request] Add GPU Version

Great work! It would be worth to add a gpu version and python3.7 🙏 🤗

Not able to infer

Python = 3.7
paddlepaddle=1.8.0

I am getting the following error:

NotImplementedError: Wrong number or type of arguments for overloaded function 'ctc_beam_search_decoder_batch'.

I suspect it is coming from swig.