sound-classificationtransformerasrspeech-synthesisvoice-cloningpunctuation-restorationstreaming-ttsspeech-recognitionvocoderkwsstreaming-asrspeech-alignmentttsconformerspeech-translationvoice-recognition
You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Xinghai Sun
8313895e85
|
8 years ago | |
---|---|---|
README.md | 8 years ago | |
audio_data_utils.py | 8 years ago | |
eng_vocab.txt | 8 years ago | |
infer.py | 8 years ago | |
librispeech.py | 8 years ago | |
model.py | 8 years ago | |
requirements.txt | 8 years ago | |
train.py | 8 years ago |
README.md
Deep Speech 2 on PaddlePaddle
Quick Start
Installation
Please replace $PADDLE_INSTALL_DIR
with your paddle installation directory.
pip install -r requirements.txt
export LD_LIBRARY_PATH=$PADDLE_INSTALL_DIR/Paddle/third_party/install/warpctc/lib:$LD_LIBRARY_PATH
For some machines, we also need to install libsndfile1. Details to be added.
Preparing Dataset(s)
python librispeech.py
More help for arguments:
python librispeech.py --help
Traininig
For GPU Training:
CUDA_VISIBLE_DEVICES=0,1,2,3 python train.py --trainer_count 4
For CPU Training:
python train.py --trainer_count 8 --use_gpu False
More help for arguments:
python train.py --help
Inferencing
python infer.py
More help for arguments:
python infer.py --help