maskgct-audio-lab / processors /text_features_extractor.py
Hecheng0625's picture
Upload 409 files
c968fc3 verified
raw
history blame
508 Bytes
# Copyright (c) 2023 Amphion.
# This source code is licensed under the MIT license found in the
# LICENSE file in the root directory of this source tree.
"""
TODO:
This module aims to be an entrance that integrates all the "text" features extraction functions.
The common text features include:
1. phone features that are used for TTS, SVS, etc.
Note:
All the features extraction are designed to utilize GPU to the maximum extent, which can ease the on-the-fly extraction for large-scale dataset.
"""