tailieunhanh - 48 Speaker Verification

Speaker recognition is the process of automatically extracting personal identity information by analysis of spoken utterances. | Furui S. Rosenberg . Speaker Verification Digital Signal Processing Handbook Ed. Vijay K. Madisetti and Douglas B. Williams Boca Raton CRC Press LLC 1999 1999 by CRC Press LLC 48 Speaker Verification Sadaoki Furui Tokyo Institute of Technology Aaron E. Rosenberg AT T Labs Research Introduction Personal Identity Characteristics Vocal Personal Identity Characteristics Basic Elements of a Speaker Recognition System Extracting Speaker Information from the Speech Signal Feature Similarity Measurements Units of Speech for Representing Speakers Input Modes Text-Dependent Fixed Passwords Text Independent No Specified Passwords Text Dependent Randomly Prompted Passwords Representations Representations That Preserve Temporal Characteristics Representations That do not Preserve Temporal Characteristics Optimizing Criteria for Model Construction Model Training and Updating Signal Feature and Score Normalization Techniques Signal Feature Normalization Likelihood and Normalized Scores Cohort or Speaker Background Models Decision Process Specifying Decision Thresholds and Measuring Performance ROC Curves Adaptive Thresholds Sequential Decisions Multi-Attempt Trials Outstanding Issues Defining Terms References Introduction Speaker recognition is the process of automatically extracting personal identity information by analysis of spoken utterances. In this section speaker recognition is taken to be a general process whereas speaker identification and speaker verification refer to specific tasks or decision modes associated with this process. Speaker identification refers to the task of determining who is speaking and speaker verification is the task of validating a speaker s claimed identity. Many applications have been considered for automatic speaker recognition. These include secure access control by voice customizing services or information to individuals by voice indexing or labeling speakers in