TUTCRIS - Tampereen teknillinen yliopisto

TUTCRIS

Exemplar-based speech enhancement for deep neural network based automatic speech recognition

Tutkimustuotosvertaisarvioitu

Yksityiskohdat

AlkuperäiskieliEnglanti
OtsikkoICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
KustantajaThe Institute of Electrical and Electronics Engineers, Inc.
Sivut4485-4489
Sivumäärä5
ISBN (painettu)9781467369978
DOI - pysyväislinkit
TilaJulkaistu - 4 elokuuta 2015
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaIEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING -
Kesto: 1 tammikuuta 19001 tammikuuta 2000

Conference

ConferenceIEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING
Ajanjakso1/01/001/01/00

Tiivistelmä

Deep neural network (DNN) based acoustic modelling has been successfully used for a variety of automatic speech recognition (ASR) tasks, thanks to its ability to learn higher-level information using multiple hidden layers. This paper investigates the recently proposed exemplar-based speech enhancement technique using coupled dictionaries as a pre-processing stage for DNN-based systems. In this setting, the noisy speech is decomposed as a weighted sum of atoms in an input dictionary containing exemplars sampled from a domain of choice, and the resulting weights are applied to a coupled output dictionary containing exemplars sampled in the short-time Fourier transform (STFT) domain to directly obtain the speech and noise estimates for speech enhancement. In this work, settings using input dictionary of exemplars sampled from the STFT, Mel-integrated magnitude STFT and modulation envelope spectra are evaluated. Experiments performed on the AURORA-4 database revealed that these pre-processing stages can improve the performance of the DNN-HMM-based ASR systems with both clean and multi-condition training.