View Full Document

EFFICIENT JOINT COMPENSATION OF SPEECH FOR THE EFFECTS OF ADDITIVE NOISE AND LINEAR FILTERING



View the full content.
View Full Document
View Full Document

8 views

Unformatted text preview:

EFFICIENT JOINT COMPENSATION OF SPEECH FOR THE EFFECTS OF ADDITIVE NOISE AND LINEAR FILTERING Fu Hua Liu1 Alejandro Acero2 and Richard M Stern1 1Department of Electrical and Computer Engineering School of Computer Science Carnegie Mellon University Pittsburgh PA 15213 ABSTRACT As automatic speech recognition systems are finding their way into practical applications it is becoming increasingly clear that they must be able to accommodate a variety of acoustical environments This paper describes two algorithms that provide robustness for automatic speech recognition systems in a fashion that is suitable for real time environmental normalization for workstations of moderate size The first algorithm is a modification of the previously described SDCN and FCDCN algorithms except that unlike these algorithms it provides computationallyefficient environmental normalization without prior knowledge of the acoustical characteristics of the environment in which the system will be operated The second algorithm is a modification of the more complex CDCN algorithm that enables it to perform environmental compensation in better than real time We compare the recognition accuracy computational complexity and amount of training data needed to adapt to new acoustical environments using these algorithms with several different types of headset mounted and desktop microphones 1 INTRODUCTION Results of several studies have demonstrated that even automatic speech recognition systems that are designed to be speaker independent can perform very poorly when they are tested using a different type of microphone or acoustical environment from the one with which they were trained e g 1 2 3 For example the recognition accuracy of the SPHINX speech recognition system on a speaker independent alphanumeric task dropped from 85 correct to less than 20 correct when the close talking Sennheiser HMD 414 microphone CLSTLK used in training was replaced by the omnidirectional Crown PZM6FS desktop microphone



Access the best Study Guides, Lecture Notes and Practice Exams

Loading Unlocking...
Login

Join to view EFFICIENT JOINT COMPENSATION OF SPEECH FOR THE EFFECTS OF ADDITIVE NOISE AND LINEAR FILTERING and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view EFFICIENT JOINT COMPENSATION OF SPEECH FOR THE EFFECTS OF ADDITIVE NOISE AND LINEAR FILTERING and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?