Dockerfile

# Dockerfile for speech to text (STT) based on Kaldi ASR (http://kaldi-asr.org/)
# Kaldi's 'chain' models (type of DNN-HMM model) used
# Model released by api.ai team
# (https://github.com/api-ai/api-ai-english-asr-model)
#
# (c) Abylay Ospan <aospan@jokersys.com>, 2017
# https://jokersys.com
# under GPLv2 license

FROM ubuntu:14.04
MAINTAINER aospan@jokersys.com

RUN apt-get update && apt-get install -y --no-install-recommends \
  build-essential cmake git wget python-dev unzip \
  python-numpy python-scipy curl python-tk libatlas3-base \
  ca-certificates zlib1g-dev automake autoconf libtool subversion && \
  curl -k https://bootstrap.pypa.io/get-pip.py  > get-pip.py && python get-pip.py && \
  rm -rf /var/lib/apt/lists/*

WORKDIR /opt

# * compile Kaldi
# * make image more compact:
#   strip binary files, remove *.o files
# * download api.ai model
RUN git clone https://github.com/aospan/kaldi.git && \
  sudo ln -s -f bash /bin/sh && \
  cd kaldi/tools/ && make -j"$(nproc)" && \
  cd ../src && ./configure --shared && \
  make depend -j"$(nproc)" && make -j"$(nproc)" && \
  find /opt/kaldi/ -name "*.o" -exec rm -f {} \; && \
  find /opt/kaldi/ -type f -perm +100 -exec strip {} \; 2>/dev/null && \
  find /opt/kaldi/ -name "*.a" -exec rm {} \; 2>/dev/null && \
  find /opt/kaldi/ -name "*.so" -exec strip {} \; 2>/dev/null && \
  cd /opt/kaldi/egs/apiai_decode/s5 && ./download-model.sh

WORKDIR /opt/kaldi
COPY entrypoint.sh /entrypoint.sh
COPY in.wav /opt/in/
CMD /entrypoint.sh