语音识别与合成的基本程序，分为特征提取、模型建立和语音识别，且包括特征补偿-Basic procedures for speech recognition and synthesis, divided into feature extraction, modeling and speech recognition, including feature compensation