experiments/STT
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Revert "Merge pull request #3246 from lissyx/fix-docker"

Browse Source 
This reverts commit c01fda56c0, reversing
changes made to 3e99b0d8b2.
This commit is contained in:
Reuben Morais 2020-08-25 15:35:03 +02:00
parent c62a604876
commit 7a6508612d
3 changed files with 16 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
Dockerfile.build.tmpl
View File

@ -114,14 +114,14 @@ RUN echo "build --spawn_strategy=standalone --genrule_strategy=standalone" \
WORKDIR / WORKDIR /
RUN git clone --recursive $MOZILLA_VOICE_STT_REPO RUN git clone --recursive $MOZILLA_VOICE_STT_REPO
WORKDIR /STT WORKDIR /DeepSpeech
RUN git checkout $MOZILLA_VOICE_STT_SHA RUN git checkout $MOZILLA_VOICE_STT_SHA
RUN git submodule sync tensorflow/ RUN git submodule sync tensorflow/
RUN git submodule update --init tensorflow/ RUN git submodule update --init tensorflow/
# >> START Build and bind # >> START Build and bind
WORKDIR /STT/tensorflow WORKDIR /DeepSpeech/tensorflow
# Fix for not found script https://github.com/tensorflow/tensorflow/issues/471 # Fix for not found script https://github.com/tensorflow/tensorflow/issues/471
RUN ./configure RUN ./configure
@ -132,7 +132,7 @@ RUN ./configure
# passing LD_LIBRARY_PATH is required cause Bazel doesn't pickup it from environment # passing LD_LIBRARY_PATH is required cause Bazel doesn't pickup it from environment
# Build Mozilla Voice STT # Build DeepSpeech
RUN bazel build \ RUN bazel build \
--workspace_status_command="bash native_client/bazel_workspace_status_cmd.sh" \ --workspace_status_command="bash native_client/bazel_workspace_status_cmd.sh" \
--config=monolithic \ --config=monolithic \
@ -153,18 +153,18 @@ RUN bazel build \
--verbose_failures \ --verbose_failures \
--action_env=LD_LIBRARY_PATH=${LD_LIBRARY_PATH} --action_env=LD_LIBRARY_PATH=${LD_LIBRARY_PATH}
# Copy built libs to /STT/native_client # Copy built libs to /DeepSpeech/native_client
RUN cp bazel-bin/native_client/libmozilla_voice_stt.so /STT/native_client/ RUN cp bazel-bin/native_client/libmozilla_voice_stt.so /DeepSpeech/native_client/
# Build client.cc and install Python client and decoder bindings # Build client.cc and install Python client and decoder bindings
ENV TFDIR /STT/tensorflow ENV TFDIR /DeepSpeech/tensorflow
RUN nproc RUN nproc
WORKDIR /STT/native_client WORKDIR /DeepSpeech/native_client
RUN make NUM_PROCESSES=$(nproc) mozilla_voice_stt RUN make NUM_PROCESSES=$(nproc) mozilla_voice_stt
WORKDIR /STT WORKDIR /DeepSpeech
RUN cd native_client/python && make NUM_PROCESSES=$(nproc) bindings RUN cd native_client/python && make NUM_PROCESSES=$(nproc) bindings
RUN pip3 install --upgrade native_client/python/dist/*.whl RUN pip3 install --upgrade native_client/python/dist/*.whl
@ -176,8 +176,8 @@ RUN pip3 install --upgrade native_client/ctcdecode/dist/*.whl
# Allow Python printing utf-8 # Allow Python printing utf-8
ENV PYTHONIOENCODING UTF-8 ENV PYTHONIOENCODING UTF-8
# Build KenLM in /STT/native_client/kenlm folder # Build KenLM in /DeepSpeech/native_client/kenlm folder
WORKDIR /STT/native_client WORKDIR /DeepSpeech/native_client
RUN rm -rf kenlm && \ RUN rm -rf kenlm && \
git clone https://github.com/kpu/kenlm && \ git clone https://github.com/kpu/kenlm && \
cd kenlm && \ cd kenlm && \
@ -188,4 +188,4 @@ RUN rm -rf kenlm && \
make -j $(nproc) make -j $(nproc)
# Done # Done
WORKDIR /STT WORKDIR /DeepSpeech

10
Dockerfile.train.tmpl
View File

@ -31,11 +31,9 @@ RUN apt-get install -y --no-install-recommends libopus0 libsndfile1
RUN rm -rf /var/lib/apt/lists/* RUN rm -rf /var/lib/apt/lists/*
WORKDIR / WORKDIR /
RUN echo git clone $MOZILLA_VOICE_STT_REPO
RUN git clone $MOZILLA_VOICE_STT_REPO RUN git clone $MOZILLA_VOICE_STT_REPO
WORKDIR /STT WORKDIR /DeepSpeech
RUN echo git checkout $MOZILLA_VOICE_STT_SHA
RUN git checkout $MOZILLA_VOICE_STT_SHA RUN git checkout $MOZILLA_VOICE_STT_SHA
# Build CTC decoder first, to avoid clashes on incompatible versions upgrades # Build CTC decoder first, to avoid clashes on incompatible versions upgrades
@ -45,7 +43,7 @@ RUN pip3 install --upgrade native_client/ctcdecode/dist/*.whl
# Prepare deps # Prepare deps
RUN pip3 install --upgrade pip==20.0.2 wheel==0.34.2 setuptools==46.1.3 RUN pip3 install --upgrade pip==20.0.2 wheel==0.34.2 setuptools==46.1.3
# Install Mozilla Voice STT # Install DeepSpeech
# - No need for the decoder since we did it earlier # - No need for the decoder since we did it earlier
# - There is already correct TensorFlow GPU installed on the base image, # - There is already correct TensorFlow GPU installed on the base image,
# we don't want to break that # we don't want to break that
@ -56,7 +54,7 @@ RUN python3 util/taskcluster.py --source tensorflow --branch r1.15 \
--artifact convert_graphdef_memmapped_format --target . --artifact convert_graphdef_memmapped_format --target .
# Build KenLM to generate new scorers # Build KenLM to generate new scorers
WORKDIR /STT/native_client WORKDIR /DeepSpeech/native_client
RUN rm -rf kenlm && \ RUN rm -rf kenlm && \
git clone https://github.com/kpu/kenlm && \ git clone https://github.com/kpu/kenlm && \
cd kenlm && \ cd kenlm && \
@ -65,6 +63,6 @@ RUN rm -rf kenlm && \
cd build && \ cd build && \
cmake .. && \ cmake .. && \
make -j $(nproc) make -j $(nproc)
WORKDIR /STT WORKDIR /DeepSpeech
RUN ./bin/run-ldc93s1.sh RUN ./bin/run-ldc93s1.sh

2
setup.py
View File

@ -98,7 +98,7 @@ def main():
setup( setup(
name='mozilla_voice_stt_training', name='mozilla_voice_stt_training',
version=version, version=version,
description='Training code for Mozilla Voice STT', description='Training code for mozilla DeepSpeech',
url='https://github.com/mozilla/STT', url='https://github.com/mozilla/STT',
author='Mozilla', author='Mozilla',
license='MPL-2.0', license='MPL-2.0',