maxmustermann/DeepLearningExamples
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity

Made processing faster.

Browse source 
No need to process full dataset. We just need sample inputs for model conversion.
This commit is contained in:
Deepak Mangla 2021-08-29 23:58:02 +05:30 committed by GitHub
parent 248927e6fd
commit e27e2277cf
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 1 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
PyTorch/SpeechSynthesis/FastPitch/triton/scripts/process_dataset.sh
View file

@ -23,9 +23,7 @@ DATASET_DIR="${DATASETS_DIR}/LJSpeech-1.1/LJSpeech-1.1_fastpitch"
if [ ! -d "${DATASET_DIR}/mels" ]; then
python extract_pitch.py \
--wav-text-filelists filelists/ljs_audio_text_train_v3.txt \
filelists/ljs_audio_text_val.txt \
filelists/ljs_audio_text_test.txt \
--wav-text-filelists filelists/ljs_audio_text_val.txt \
--n-workers 16 \
--batch-size 1 \
--dataset-path $DATASET_DIR \