Build your own voice
To build your own voice,
cd egs/build_your_own_voice/s1 and follow the below steps:
The first step is to run setup as it creates directories and some text files for testing.
The next steps demonstrate on how to setup voice.
It also creates a global config file:
conf/global_settings.cfg, where default settings are stored.
You need to modify these params as per your own data.
To prepare labels
./02_prepare_labels.sh <path_to_wav_dir> <path_to_text_dir> <path_to_labels_dir>
Prepare acoustic features
To prepare acoustic features
./03_prepare_acoustic_features.sh <path_to_wav_dir> <path_to_feat_dir>
Prepare config files
At this point, we have to prepare two config files to train DNN models - Acoustic Model - Duration Model
To prepare config files:
Four config files will be generated: two for training, and two for testing.
Train duration model
To train duration model:
Train acoustic model
To train acoustic model:
To synthesize speech:
./07_run_merlin.sh <path_to_text_dir> <path_to_test_dur_conf_file> <path_to_test_synth_conf_file>