File size: 664 Bytes
901e06a | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 | lang=$1
CVSS_ROOT=/data/zhangshaolei/datasets/cvss/cvss-c
COVOST2_ROOT=/data/zhangshaolei/datasets/covost2
ROOT=/data/zhangshaolei/StreamSpeech
PRETRAIN_ROOT=$ROOT/pretrain_models
PYTHONPATH=$ROOT/fairseq python $ROOT/preprocess_scripts/prep_cvss_c_multilingual_data.py \
--covost-data-root $COVOST2_ROOT/ --cvss-data-root $CVSS_ROOT/ \
--output-root $CVSS_ROOT/$lang-en \
--src-lang $lang \
--target-type unit --unit-type km1000 --reduce-unit \
--vocoder-checkpoint $PRETRAIN_ROOT/unit-based_HiFi-GAN_vocoder/mHuBERT.layer11.km1000.en/g_00500000 --vocoder-cfg $PRETRAIN_ROOT/unit-based_HiFi-GAN_vocoder/mHuBERT.layer11.km1000.en/config.json
|