Share on Facebook Tweet on Twitter Share on LinkedIn Share by email
Voice Source Estimation for Artificial Bandwidth Extension of Telephone Speech

M. R. P. Thomas, B. Geiser, J. Gudnason, P. A. Naylor, and P. Vary

Abstract

Artificial bandwidth extension (ABWE) of speech signals aims to estimate wideband speech (50 Hz – 7 kHz) from narrowband signals (300 Hz – 3.4 kHz). Applying the source-filter model of speech, many existing algorithms estimate vocal tract filter parameters independently of the source signal. However, many current methods for extending the narrowband voice source signal are limited to straightforward signal processing techniques which are only effective for high-band estimation. This paper presents a method for ABWE that employs novel data-driven modelling and an existing spectral mirroring technique to estimate the wideband source signal in both the high and low extension bands. A state-of the-art Hidden Markov Model-based estimator evaluates the temporal and spectral envelopes in the missing frequency bands, with which the ABWE speech signal is synthesized. Informal listening tests comparing two existing source estimation techniques and two permutations of the proposed approach show an improvement in the perceived bandwidth of speech signals, in particular towards low frequencies. Subjective tests on the same data show a preference for the proposed techniques over the existing methods under test.

Details

Publication typeInproceedings
Published inProc. IEEE Intl. Conf. on Acoustics, Speech and Signal Processing (ICASSP)
AddressDallas, TX, USA
> Publications > Voice Source Estimation for Artificial Bandwidth Extension of Telephone Speech