SPEECH RECOGNITION AS FEATURE EXTRACTION FOR SPEAKER RECOGNITION

Information from speech recognition can be used in various ways in state-of-the-art speaker recognition systems. This includes the obvious use of recognized words to enable the use of text-dependent speaker modeling techniques when the words spoken are not given. Furthermore, it has been shown that the choice of words and phones itself can be a useful indicator of speaker identity. Also, recognizer output enables higher-level features, in particular those related to prosodic properties of speech. Finally, we discuss the use of mere byproducts of word recognition, such as subword unit alignments, pronunciations, and speaker adaptation transforms to derive powerful nonstandard features for speaker modeling. We present specific techniques and results from SRI’s NIST speaker recognition evaluation system. Index Terms — Speaker recognition, speech recognition, high-level features, speaker adaptation, prosody. 1