ISCA Archive Interspeech 2008
ISCA Archive Interspeech 2008

Front-end for far-field speech recognition based on frequency domain linear prediction

Sriram Ganapathy, Samuel Thomas, Hynek Hermansky

Automatic Speech Recognition (ASR) systems usually fail when they encounter speech from far-field microphone in reverberant environments. This is due to the application of short-term feature extraction techniques which do not compensate for the artifacts introduced by long room impulse responses. In this paper, we propose a front-end, based on Frequency Domain Linear Prediction (FDLP), that tries to remove reverberation artifacts present in far-field speech. Long temporal segments of far-field speech are analyzed in narrow frequency sub-bands to extract FDLP envelopes and residual signals. Filtering the residual signals with gain normalized inverse FDLP filters result in a set of sub-band signals which are synthesized to reconstruct the signal back. ASR experiments on far-field speech data processed by the proposed front-end show significant improvements (relative reduction of 30% in word error rate) compared to other robust feature extraction techniques.


doi: 10.21437/Interspeech.2008-287

Cite as: Ganapathy, S., Thomas, S., Hermansky, H. (2008) Front-end for far-field speech recognition based on frequency domain linear prediction. Proc. Interspeech 2008, 984-987, doi: 10.21437/Interspeech.2008-287

@inproceedings{ganapathy08b_interspeech,
  author={Sriram Ganapathy and Samuel Thomas and Hynek Hermansky},
  title={{Front-end for far-field speech recognition based on frequency domain linear prediction}},
  year=2008,
  booktitle={Proc. Interspeech 2008},
  pages={984--987},
  doi={10.21437/Interspeech.2008-287}
}