EUROSPEECH 2001 Scandinavia
This paper presents the robust front-end algorithm that was submitted by Motorola to the ETSI STQ-Aurora DSR working group as a proposal for the Advanced DSR front-end in January 2001. The algorithm consists of a two-stage mel-warped Wiener filter, a waveform processor, a channel-normalized mel-frequency cepstral calculation and a subsystem of post-cepstral processing according to the reliability of mel-spectral components, etc. The output of this algorithm, a set of Mel-Frequency Cepstral Coefficients (MFCC), is compressed, encoded and transmitted at 4800 bps. Compared to ETSI standard MFCC front-end, the proposed algorithm delivers an improvement of 47.58% on the Aurora 2 database, which is required by this Eurospeech special event. In this paper we also give further insights about the proposal by providing performances and analyses with the Aurora SpeechDat-Car databases.
Bibliographic reference. Cheng, Yan Ming / Macho, Dusan / Wei, Yuanjun / Ealey, Douglas / Kelleher, Holly / Pearce, David / Kushner, William / Ramabadran, Tenkasi (2001): "A robust front-end algorithm for distributed speech recognition", In EUROSPEECH-2001, 425-428.