语音识别与语者识别基础技术.doc

语音识别与语者识别基础技术
语音讯号处理基本观念
语者识别基础技术
语音识别基础技术
教学成果及未来工作方向
中华技术学院电子系所
陈梧桐副教授
语音讯号处理基本观念
语音信号的产生(Speech Production Mechanism)
本图引用自Thomas F. Quatieri, Discrete-time speech signal processing principles and practice, Prentice Hall PTR, 2002
语音讯号的分类(Categorization of Speech Sounds)
本图引用自Thomas F. Quatieri, Discrete-time speech signal processing principles and practice, Prentice Hall PTR, 2002
本图引用自Thomas F. Quatieri, Discrete-time speech signal processing principles and practice, Prentice Hall PTR, 2002
数字语音讯号处理
本图引用自Thomas F. Quatieri, Discrete-time speech signal processing principles and practice, Prentice Hall PTR, 2002
离散时间讯号(Discrete-Time Signals)
脉冲(Impulse)讯号
离散时间系统(Discrete-Time Systems)
Convolution Model
,
其中称为脉冲响应(Impulse Response)
离散时间富利叶转换(Discrete-Time Fourier Transform)

例1.

Z-转换(Z-Transform)

频域下的线性非时变系统(LTI Systems in the Frequency Domain)
Convolution定理(Convolution Theorem)


Windowing定理(Windowing Theorem)


例2.
()
本图引用自Thomas F. Quatieri, Discrete-time speech signal processing principles and practice, Prentice Hall PTR, 2002
例3.
本图引用自Thomas F. Quatieri, Discrete-time speech signal processing principles and practice, Prentice Hall PTR, 2002