DeepSpeech/.compute

#!/bin/bash

set -xe

apt-get install -y python3-venv libopus0

python3 -m venv /tmp/venv
source /tmp/venv/bin/activate

pip install -U setuptools wheel pip
pip install .
pip uninstall -y tensorflow
pip install tensorflow-gpu==1.14

mkdir -p ../keep/summaries

data="${SHARED_DIR}/data"
fis="${data}/LDC/fisher"
swb="${data}/LDC/LDC97S62/swb"
lbs="${data}/OpenSLR/LibriSpeech/librivox"
cv="${data}/mozilla/CommonVoice/en_1087h_2019-06-12/clips"
npr="${data}/NPR/WAMU/sets/v0.3"

python -u DeepSpeech.py \
  --train_files "${npr}/best-train.sdb","${npr}/good-train.sdb","${cv}/train.sdb","${fis}-train.sdb","${swb}-train.sdb","${lbs}-train-clean-100.sdb","${lbs}-train-clean-360.sdb","${lbs}-train-other-500.sdb" \
  --dev_files "${lbs}-dev-clean.sdb" \
  --test_files "${lbs}-test-clean.sdb" \
  --train_batch_size 24 \
  --dev_batch_size 48 \
  --test_batch_size 48 \
  --train_cudnn \
  --n_hidden 2048 \
  --learning_rate 0.0001 \
  --dropout_rate 0.40 \
  --epochs 150 \
  --noearly_stop \
  --feature_cache "../tmp/feature.cache" \
  --checkpoint_dir "../keep" \
  --summary_dir "../keep/summaries"