Speaker normalization for automatic speech recognition — An on-line approach

We propose a method to transform the on line speech signal so as to comply with the specifications of an HMM-based automatic speech recognizer. The spectrum of the input signal undergoes a vocal tract length (VTL) normalization based on differences of the average third formant F3. The high frequency gap which is generated after scaling is estimated by means of an extrapolation scheme. Mel scale cepstral coefficients (MFCC) are used along with delta and delta
²
-cepstra as well as delta and delta
²
energy. The method has been tested on the TI digits database which contains adult and kids speech providing substantial gains with respect to non normalized speech.