Improving recognition of proper nouns in ASR through generating and filtering phonetic transcriptions

Accurate phonetic transcription of proper nouns can be an important resource for commercial applications that embed speech echnologies, such as audio indexing and vocal phone directory lookup. However, an accurate phonetic transcription is more difficult o obtain for proper nouns than for regular words. Indeed, phonetic transcription of a proper noun depends on both the origin of the peaker pronouncing it and the origin of the proper noun itself. This work proposes a method that allows the extraction of phonetic transcriptions of proper nouns using actual utterances of those roper nouns, thus yielding transcriptions based on practical use instead of mere pronunciation rules. The proposed method consists in a process that first extracts phonetic transcriptions, and then iteratively filters them. In order o initialize the process, an alignment dictionary is used to detect word boundaries. A rule-based grapheme-to-phoneme generator LIA PHON), a knowledge-based approach (JSM), and a Statistical Machine Translation based system were evaluated for this lignment. As a result, compared to our reference dictionary (BDLEX supplemented by LIA PHON for missing words) on the STER 1 French broadcast news corpus, we were able to significantly decrease the Word Error Rate (WER) on segments of speech ith proper nouns, without negatively affecting the WER on the rest of the corpus. 2014 Elsevier Ltd. All rights reserved.