EUROSPEECH 2001 Scandinavia
7th European Conference on Speech Communication and Technology
2nd INTERSPEECH Event

Aalborg, Denmark
September 3-7, 2001

                 

A New Method for Speech Recognition in the Presence of Non-stationary, Unpredictable and High-level Noise

Ikuyo Masuda-Katsuse

Kyushu Institute of Design / Institute of Systems & Information Technologies, Japan

We propose a new method for speech recognition in the presence of non-stationary, unpredictable and high-level noise by extending PreFEst (Predominant-F0 Estimation Method) which was developed to estimate melody and bass lines from music signals. The proposed method does not need to know about noise characteristics in advance and does not even estimate them in its process. A small set of evaluations demonstrates the feasibility of the method by showing a good performance even when the background noise is real and non-stationary noise and the SNR is less than 10 dB.

Full Paper

Bibliographic reference.  Masuda-Katsuse, Ikuyo (2001): "A new method for speech recognition in the presence of non-stationary, unpredictable and high-level noise", In EUROSPEECH-2001, 1119-1122.