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

Merge pull request #3108 from DanBmh/update_docker

Browse Source 
Build kenlm in training container image.
This commit is contained in:
lissyx 2020-06-26 16:17:44 +02:00 committed by GitHub
commit db717d7f73
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 14 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
Dockerfile.train.tmpl
View File

@ -10,9 +10,11 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
apt-utils \ apt-utils \
bash-completion \ bash-completion \
build-essential \ build-essential \
cmake \
curl \ curl \
git \ git \
git-lfs \ git-lfs \
libboost-all-dev \
libbz2-dev \ libbz2-dev \
locales \ locales \
python3-venv \ python3-venv \
@ -50,4 +52,16 @@ RUN DS_NODECODER=y DS_NOTENSORFLOW=y pip3 install --upgrade -e .
RUN python3 util/taskcluster.py --source tensorflow --branch r1.15 \ 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
WORKDIR /DeepSpeech/native_client
RUN rm -rf kenlm && \
git clone https://github.com/kpu/kenlm && \
cd kenlm && \
git checkout 87e85e66c99ceff1fab2500a7c60c01da7315eec && \
mkdir -p build && \
cd build && \
cmake .. && \
make -j $(nproc)
WORKDIR /DeepSpeech
RUN ./bin/run-ldc93s1.sh RUN ./bin/run-ldc93s1.sh