Parameter optimization for vocal tract length normalization

摘要: This paper focuses on the optimization of model parameters for vocal tract length normalization (VTLN). For maximum likelihood (ML) based techniques, complexity VTL-models is a source variation in system performance. An optimal VTL-model that ensures best global word error rate proposed. The choice frequency warping factor also depends signal processing step VTLN. A set VTLN stage proposed with extensive results an range.