VTLN based on the linear interpolation of contiguous mel filter-bank energies

This paper describes a novel feature-space VTLN method that models frequency warping as a linear interpolation of contiguous Mel filter-bank energies. The presented technique aims to reduce the distortion in the Mel filter-bank energy estimation due to the harmonic composition of voiced speech intervals and DFT sampling when the central frequency of band-pass filters is shifted. The presented interpolated filterbank energy-based VTLN leads to relative reductions in WER as high as 11.2% and 7.6% when compared with the baseline system and standard VTLN, respectively, in a mediumvocabulary continuous speech recognition task. Also, this new scheme provides significant reductions in WER equal to 7% when compared with state-of-the-art VTLN methods based on linear transforms in the cepstral space. The warping factor estimated here shows more dependence on the speaker and more independence of the acoustic-phonetic content than the warping factor in state-of-the-art VTLN techniques.