NeRF/data_utils/deepspeech_features/README.md

Routines for DeepSpeech features processing

Several routines for DeepSpeech features processing, like speech features generation for VOCA model.

Installation

pip3 install -r requirements.txt

Usage

Generate wav files:

python3 extract_wav.py --in-video=<you_data_dir>

Generate files with DeepSpeech features:

python3 extract_ds_features.py --input=<you_data_dir>