Large Vocabulary Continuous Speech Recognition (LVCSR) systems decode the input speech using diverse information sources, such as acoustic, lexical, and linguistic. Although most of the unreliable hypotheses are pruned during the recognition process, current state-of-the-art systems often make errors that are “unreasonable” for human listeners. Several studies have shown that a proper integration of acoustic-phonetic information can be beneficial to reducing such errors. We have previously shown that high-accuracy phone recognition can be achieved if a bank of speech attribute detectors is used to compute a confidence score describing attribute activation levels that the current frame exhibits. In those experiments, the phone recognition system did not rely on the language model to follow their word sequence constraints, and the vocabulary was small. In this work, we extend our approach to LVCSR by introducing a second recognition step during which additional information not directly used during conventional log-likelihood based decoding is introduced. Experimental results show promising performance.

SINISCALCHI, S.M., Torbjorn Svendsen, Chin Hui Lee (2009). A phonetic feature based lattice rescoring approach to LVCSR. In IEEE ICASSP (pp. 3865-3868) [10.1109/ICASSP.2009.4960471].

A phonetic feature based lattice rescoring approach to LVCSR

SINISCALCHI, SABATO MARCO
Primo
Formal Analysis
;
2009-01-01

Abstract

Large Vocabulary Continuous Speech Recognition (LVCSR) systems decode the input speech using diverse information sources, such as acoustic, lexical, and linguistic. Although most of the unreliable hypotheses are pruned during the recognition process, current state-of-the-art systems often make errors that are “unreasonable” for human listeners. Several studies have shown that a proper integration of acoustic-phonetic information can be beneficial to reducing such errors. We have previously shown that high-accuracy phone recognition can be achieved if a bank of speech attribute detectors is used to compute a confidence score describing attribute activation levels that the current frame exhibits. In those experiments, the phone recognition system did not rely on the language model to follow their word sequence constraints, and the vocabulary was small. In this work, we extend our approach to LVCSR by introducing a second recognition step during which additional information not directly used during conventional log-likelihood based decoding is introduced. Experimental results show promising performance.
2009
Settore IINF-05/A - Sistemi di elaborazione delle informazioni
978-1-4244-2354-5
SINISCALCHI, S.M., Torbjorn Svendsen, Chin Hui Lee (2009). A phonetic feature based lattice rescoring approach to LVCSR. In IEEE ICASSP (pp. 3865-3868) [10.1109/ICASSP.2009.4960471].
File in questo prodotto:
File Dimensione Formato  
04960471.pdf

Solo gestori archvio

Dimensione 253.26 kB
Formato Adobe PDF
253.26 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10447/670047
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 9
  • ???jsp.display-item.citation.isi??? 3
social impact