Interspeech'2005 - Eurospeech

Lisbon, Portugal
September 4-8, 2005

Bandwidth Expansion of Narrowband Speech Using Non-Negative Matrix Factorization

Dhananjay Bansal (1), Bhiksha Raj (2), Paris Smaragdis (2)

(1) Convergys Corporation, USA; (2) Mitsubishi Electric Research Laboratories, USA

In this paper, we present a novel technique for the estimation of the high frequency components (4-8kHz) of speech signals from narrow-band (0-4 kHz) signals using convolutive Non-Negative Matrix Factorisation (NMF). The proposed technique utilizes a brief recording of simultaneous broad band and narrow band signals from a target speaker to learn a set of broad-band non-negative "bases" for the speaker. The low-frequency components of these bases are used to determine how the high-frequency components must be combined in order to reconstruct the high-frequency components of new narrow-band signals from the speaker. Experiments reveal that the technique is able to reconstruct broadband speech that is perceptually virtually indistinguishable from true broadband recordings.

Full Paper

Bibliographic reference.  Bansal, Dhananjay / Raj, Bhiksha / Smaragdis, Paris (2005): "Bandwidth expansion of narrowband speech using non-negative matrix factorization", In INTERSPEECH-2005, 1505-1508.