Pronunciation-based ASR for names

To improve the ASR of proper names a novel method based on the generation of pronunciation variants by means of phoneme-to-phoneme converters (P2Ps) is proposed. The aim is to convert baseline transcriptions into variants that maximally resemble actual name pronunciations that were found in a training corpus. The method has to operate in a cross lingual setting with native Dutch persons speaking Dutch and foreign names, and foreign persons speaking Dutch names. The P2Ps are trained to act either on conventional G2P-transcriptions or on canonical transcriptions that were provided by a human expert. Including the variants produced by the P2Ps in the lexicon of the recognizer substantially improves the recognition accuracy for natives pronouncing foreign names, but not for the other investigated combinations.

[1]  Louis Boves,et al.  The Onomastica interlanguage pronunciation lexicon , 1995, EUROSPEECH.

[2]  Jean-Pierre Martens,et al.  Recognition of foreign names spoken by native speakers , 2007, INTERSPEECH.

[3]  Bart D'hoore,et al.  How speaker tongue and name source language affect the automatic recognition of spoken names , 2009, INTERSPEECH.

[4]  Stefan Schaden,et al.  Regelbasierte Modellierung fremdsprachlich akzentbehafteter Aussprachevarianten. Grundlagen, Entwurf und Implementierung eines Regelsystems für sprachtechnologische Anwendungen , 2006 .

[5]  Bart D'hoore,et al.  The AUTONOMATA Spoken Names Corpus , 2008, LREC.

[6]  Qian Yang,et al.  Development of a phoneme-to-phoneme (p2p) converter to improve the grapheme-to-phoneme (g2p) conversion of names , 2006, LREC.

[7]  Mitch Weintraub,et al.  Learning linguistically valid pronunciations from acoustic data , 2003, INTERSPEECH.

[8]  Gosse Bouma,et al.  A Finite State and Data-Oriented Method for Grapheme to Phoneme Conversion , 2000, ANLP.

[9]  Paul Taylor,et al.  Hidden Markov models for grapheme to phoneme conversion , 2005, INTERSPEECH.

[10]  Ariadna Font Llitjós,et al.  Evaluation and collection of proper name pronunciations online , 2002, LREC.

[11]  Hermann Ney,et al.  Multigram-based grapheme-to-phoneme conversion for LVCSR , 2003, INTERSPEECH.

[12]  Christophe d'Alessandro,et al.  Evaluating the pronunciation of proper names by four French grapheme-to-phoneme converters , 2005, INTERSPEECH.