diff --git a/.compute b/.compute old mode 100755 new mode 100644 index a18305f0..7e4ff5d0 --- a/.compute +++ b/.compute @@ -2,24 +2,33 @@ set -xe -pip3 install -r <(grep -v tensorflow requirements.txt) -pip3 install tensorflow-gpu==1.13.0-rc2 +apt-get install -y python3-venv swig3.0 +ln -s /usr/bin/swig3.0 /usr/bin/swig + +python3 -m venv /tmp/venv +source /tmp/venv/bin/activate + +pip install -r <(grep -v tensorflow requirements.txt) +pip install tensorflow-gpu==1.13.0-rc2 + +pushd native_client/ctcdecode +make clean +make NUM_PROCESSES=16 +pip install dist/*.whl +popd # Install ds_ctcdecoder package from TaskCluster -pip3 install $(python3 util/taskcluster.py --decoder) +# pip install $(python3 util/taskcluster.py --decoder) mkdir -p ../keep/summaries -data="${SHARED_DIR}/data" -fis="${data}/LDC/fisher" -swb="${data}/LDC/LDC97S62/swb" -lbs="${data}/OpenSLR/LibriSpeech/librivox" -cv="${data}/mozilla/CommonVoice/v2.0-alpha2.0/en/cv_en_valid" +LANG="sl" +cv="$SHARED_DIR/data/mozilla/CommonVoice/v2.0-alpha2.0/$LANG" -python3 -u DeepSpeech.py \ - --train_files '/home/josh/Downloads/sl/train.csv' \ - --dev_files '/home/josh/Downloads/sl/dev.csv' \ - --test_files '/home/josh/Downloads/sl/test.csv' \ +python -u DeepSpeech.py \ + --train_files '$CV/cv_${LANG}_valid_train.csv' \ + --dev_files '$CV/cv_${LANG}_valid_dev.csv' \ + --test_files '$CV/cv_${LANG}_valid_test.csv' \ --train_batch_size 24 \ --dev_batch_size 48 \ --test_batch_size 48 \ @@ -27,7 +36,7 @@ python3 -u DeepSpeech.py \ --n_hidden 2048 \ --learning_rate 0.0001 \ --dropout_rate 0.15 \ - --epoch 30 \ + --epoch 1 \ --display_step 0 \ --validation_step 1 \ --checkpoint_dir "../keep" \ diff --git a/.install b/.install deleted file mode 100755 index dc22af30..00000000 --- a/.install +++ /dev/null @@ -1,19 +0,0 @@ -#!/bin/bash - -virtualenv -p python3 ../tmp/venv -source ../tmp/venv/bin/activate -pip install -r <(grep -v tensorflow requirements.txt) -pip install tensorflow - -python3 util/taskcluster.py --arch cpu --target ../tmp/native_client - -# Install ds_ctcdecoder package from TaskCluster -# pip install $(python3 util/taskcluster.py --decoder) - -pushd native_client/ctcdecode -make clean -make NUM_PROCESSES=16 -pip install dist/*.whl -popd - -mkdir -p ../keep/summaries diff --git a/data/alphabet.txt b/data/alphabet.txt new file mode 100644 index 00000000..282e9d85 --- /dev/null +++ b/data/alphabet.txt @@ -0,0 +1,64 @@ +## The following unique characters were found in your transcripts: ### +? +i +V +Š +Z +j +; +L +z +. +Č +r +– +o +P +H +J + +, +b +N +R +u +ž +: +! +Ž +a +p +s +G +l +š +B +F +f +c +T +M +- +v +W +e +S +y +m +t +A +g +4 +I +n +U +E +C +D +K +č +k +h +O +d +## diff --git a/data/lm/lm.binary b/data/lm/lm.binary index 0c53ca35..5de5333a 100644 --- a/data/lm/lm.binary +++ b/data/lm/lm.binary @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:e1fa6801b25912a3625f67e0f6cafcdacb24033be9fad5fa272152a0828d7193 -size 1800894585 +oid sha256:c983a78da300938a36b5ec92d6dfa5e1f19eb510de81d76ad549df245e26f0a9 +size 65714 diff --git a/data/lm/trie b/data/lm/trie deleted file mode 100644 index a937a226..00000000 --- a/data/lm/trie +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a5324f06b27c7b4ef88dd2c8bc3d05d4c718db219c2e007f59061a30c9ac7afa -size 21627983 diff --git a/data/lm/trie_utf8 b/data/lm/trie_utf8 index 6c1f103c..edf9808c 100644 --- a/data/lm/trie_utf8 +++ b/data/lm/trie_utf8 @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:144c5bfd94348bf382b700836ffd650b57fa146f7e3f739806e64509303fa2c7 -size 21628171 +oid sha256:c5aae9eb4f8b641044c8bab723cde117c123af8e1314d609bd314ef5f13569f7 +size 87211 diff --git a/native_client/ctcdecode/common.a b/native_client/ctcdecode/common.a index 3800bde3..a2c18cf6 100644 Binary files a/native_client/ctcdecode/common.a and b/native_client/ctcdecode/common.a differ diff --git a/native_client/ctcdecode/dist/ds_ctcdecoder-0.5.0a1-cp36-cp36m-linux_x86_64.whl b/native_client/ctcdecode/dist/ds_ctcdecoder-0.5.0a1-cp36-cp36m-linux_x86_64.whl index 20a40a1e..fa73ac04 100644 Binary files a/native_client/ctcdecode/dist/ds_ctcdecoder-0.5.0a1-cp36-cp36m-linux_x86_64.whl and b/native_client/ctcdecode/dist/ds_ctcdecoder-0.5.0a1-cp36-cp36m-linux_x86_64.whl differ