Evaluation of a Speech Bandwidth Extension Algorithm Based on Vocal Tract Shape Estimation

Abstract:
In this paper, we evaluate a speech bandwidth extension (BWE) algorithm which involves phonetic and speaker dependent estimation of the high-band part of the spectral envelope. The BWE algorithm extracts speech phoneme information by using a hidden Markov model. Speaker vocal tract shape information corresponding to the wideband signal is extracted by a codebook search. Postprocessing of the estimated vocal tract shape using iterative tuning allows artifacts reduction in cases of erroneous estimation of speech phoneme or vocal tract shape. We present objective measurements results demonstrating the benefit of the iterative tuning. Subjective listening tests illustrate improved wideband quality in comparison to the input narrowband speech. The algorithm complexity is also analyzed. Index Terms— Bandwidth extension, speech processing, vocal tract area function, sensitivity function, MUSHRA.