A quick hack to align single-word audio recordings
使用特定于 shell 的命令激活虚拟环境:
$ source ./venv/bin/activate # sh, bash, ksh, or zsh
$ python tf1.5/tensorflow/tensorflow/examples/speech_commands/train.py --data_dir=yj --wanted_words=up,down --data_url=
~/tf1.5/tensorflow$ python tensorflow/examples/speech_commands/freeze.py --start_checkpoint=/tmp/speech_commands_train/conv.ckpt-18000 --output_file=/tmp/yj_frozen_graph.pb --wanted_words=up,down
python tensorflow/examples/speech_commands/label_wav.py
–graph=/tmp/yj_frozen_graph.pb
–labels=/tmp/speech_commands_train/conv_labels.txt
–wav=/home/leon/yj/up/up002.wav