Autore: Franz Gerl

Current speech recognition systems are based on speaker independent speech models and suffer from inter-speaker variations in speech signal characteristics. This work develops an integrated approach …