Self-learning speaker adaptation based on spectral bias source decomposition, using very short calibration speech

摘要

A speaker adaptation technique based on the separation of speech spectra variation sources is developed for improving speaker- independent continuous speech recognition. The variation sources include speaker acoustic characteristics, and contextual dependency of allophones. Statistical methods are formulated to normalize speech spectra based on speaker acoustic characteristics and then adapt mixture Gaussian density phone models based on speaker phonologic characteristics. Adaptation experiments using short calibration speech (5 sec./speaker) have shown substantial performance improvement over the baseline recognition system.