This paper investigates methods for coping with out-of-vocabulary words in a large vocabulary speech recognition task, namely the automatic transcription of Italian broadcast news. Two alternative ways for augmenting a 64K(thousand)-word recognition vocabulary and language model are compared: introducing extra words with their phonetic transcription up to 1.2M (million) words, or extending the language model with so-called graphones, i.e. sub-word units made of phone-character sequences. Graphones and phonetic transcriptions of words are automatically generated by adapting an off-the-shelf statistical machine translation toolkit. We found that the word-based and graphone-based extensions allow both for better recognition performance, with the former performing significantly better than the latter. In addition, the word-based extension approach shows interesting potential even under conditions of little supervision. In fact, by training the grapheme to phoneme translation system with only 2K manually verified transcriptions, the final word error rate increases by just 3% relative, with respect to starting from a lexicon of 64K.
Coping with out-of-vocabulary words: open versus huge vocabulary ASR
Gerosa, Matteo;Federico, Marcello
2009-01-01
Abstract
This paper investigates methods for coping with out-of-vocabulary words in a large vocabulary speech recognition task, namely the automatic transcription of Italian broadcast news. Two alternative ways for augmenting a 64K(thousand)-word recognition vocabulary and language model are compared: introducing extra words with their phonetic transcription up to 1.2M (million) words, or extending the language model with so-called graphones, i.e. sub-word units made of phone-character sequences. Graphones and phonetic transcriptions of words are automatically generated by adapting an off-the-shelf statistical machine translation toolkit. We found that the word-based and graphone-based extensions allow both for better recognition performance, with the former performing significantly better than the latter. In addition, the word-based extension approach shows interesting potential even under conditions of little supervision. In fact, by training the grapheme to phoneme translation system with only 2K manually verified transcriptions, the final word error rate increases by just 3% relative, with respect to starting from a lexicon of 64K.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
