Automatic learning of phonetic mappings for cross-language phonetic-search in keyword spotting

Bar-Yosef, Yossi; Aloni-Lavi, Ruth; Opher, Irit; Lotner, Noam; Tetariy, Ella; Silber-Varod, Vered; Aharonson, Vered; Moyal, Ami

doi:10.1109/eeei.2012.6376955

Cited by 3 publications

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[10][11][12] The studies vary in both method for modeling [13][14][15][16] and in the mapping techniques employed. [17][18][19][20][21] Studies also differ in the amount of source LRs available, as well as, the languages concerned. A comparison of the recognition performance between the various studies reported in the literature is difficult to assess due to the differences in data and methods used.…”

Section: Introductionmentioning

confidence: 99%

Cross-language phoneme mapping for phonetic search keyword spotting using multiple source languages

Tetariy

Bar-Yosef

Gishri

et al. 2016

AIR

Self Cite

View full text Add to dashboard Cite

Performing Phonetic Search Keyword Spotting (PS KWS) in new languages when language resources are scarce is an interesting and challenging task. In a previous paper we reported a methodology that enabled PS KWS under these conditions utilizing cross-language phoneme mappings from another sufficiently resourced and well-trained source language. We performed phoneme recognition in the new target language with the acoustic model of the source language. The keyword search was performed over a phoneme lattice of the target language phonemes following a mapping from one language to the other. In the present work we extend this method and its capabilities by mapping two source language phoneme sets into one target language set and performing a combined lattice search. Testing the technique on English and Arabic as source languages yielded a 50% Detection Rate (DR) and a False Alarm Rate (FAR-measured in number of false alarms per hour per keyword) of 2 when Spanish was the target language, a DR of 36% and FAR of 4 when Dari was the target language and a DR of 35% and FAR of 6 with Farsi as the target language. These results indicate that combining two source languages is better than using a single language since the acoustic space is better represented. Searching in a combined lattice while employing adequate phoneme transformations significantly improves performance. Such a system can be used as an initial version of a PS KWS system in a new language when sufficient language resources are not available.

show abstract