I train the model by following instruction in the DeepSpeech documentation https://deepspeech.readthedocs.io/en/latest/TRAINING.html
Here is the script I use:
python DeepSpeech.py \
--train_files data/train.csv \
--dev_files data/dev.csv \
--test_files data/test.csv \
--scorer models/kenlm.scorer \
--alphabet_config_path data/alphabet.txt \
--test_batch_size 32 \
--train_batch_size 32 \
--dev_batch_size 32 \
--epochs 100 \
--early_stop True \
--es_epochs 7 \
--reduce_lr_on_plateau True \
--plateau_epochs 3 \
--force_initialize_learning_rate True \
--learning_rate 0.0001 \
--dropout_rate 0.25 \
--use_allow_growth \
--drop_source_layers 0 \
--train_cudnn \
--export_dir test/test-exp/ \
--checkpoint_dir test/test-ckp/ \
--summary_dir test/test-sum \
--max_to_keep 3 \
--augment volume[p=0.1,dbfs=-10:-40] \
--augment pitch[p=0.1,pitch=1.1~0.9] \
--augment tempo[p=0.1,factor=1.25~0.75] \
--augment dropout[p=0.1,rate=0.05] \
--augment add[p=0.1,domain=signal,stddev=0~0.5] \
--augment multiply[p=0.1,domain=features,stddev=0~0.5] \
--augment frequency_mask[p=0.1,n=1:3,size=1:5] \
--augment time_mask[p=0.1,domain=signal,n=3:10~2,size=50:100~40] \
--augment reverb[p=0.1,delay=50.0~30.0,decay=10.0:2.0~1.0] \
--augment resample[p=0.1,rate=12000:8000~4000] \
--augment codec[p=0.1,bitrate=48000:16000] \
--augment overlay[p=0.3,source=data/train.csv,layers=10:1,snr=50:20~9]
Error message:
no matches found: volume[p=0.1,dbfs=-10:-40]
The script works well if all the --augment flags are removed.
How can I solve this, PLEASE HELP.