Title :
Using web text to improve keyword spotting in speech
Author :
Gandhe, Ankur ; Long Qin ; Metze, Florian ; Rudnicky, Alex ; Lane, Ian ; Eck, Matthias
Abstract :
For low resource languages, collecting sufficient training data to build acoustic and language models is time consuming and often expensive. But large amounts of text data, such as online newspapers, web forums or online encyclopedias, usually exist for languages that have a large population of native speakers. This text data can be easily collected from the web and then used to both expand the recognizer´s vocabulary and improve the language model. One challenge, however, is normalizing and filtering the web data for a specific task. In this paper, we investigate the use of online text resources to improve the performance of speech recognition specifically for the task of keyword spotting. For the five languages provided in the base period of the IARPA BABEL project, we automatically collected text data from the web using only Limited LP resources. We then compared two methods for filtering the web data, one based on perplexity ranking and the other based on out-of-vocabulary (OOV) word detection. By integrating the web text into our systems, we observed significant improvements in keyword spotting accuracy for four out of the five languages. The best approach obtained an improvement in actual term weighted value (ATWV) of 0.0424 compared to a baseline system trained only on LimitedLP resources. On average, ATWV was improved by 0.0243 across five languages.
Keywords :
Internet; Web sites; information retrieval; natural language processing; speech recognition; text analysis; ATWV; IARPA BABEL project; LimitedLP resources; OOV; Web forums; Web text; acoustic models; actual term weighted value; keyword spotting accuracy; language models; low resource languages; native speakers; online encyclopedias; online newspapers; online text resources; out-of-vocabulary word detection; perplexity ranking; recognizer vocabulary; speech recognition; text data; Data models; Filtering; Internet; Speech; Speech recognition; Training data; Vocabulary; data filtering; keyword spotting; language modeling; low resource; web text;
Conference_Titel :
Automatic Speech Recognition and Understanding (ASRU), 2013 IEEE Workshop on
Conference_Location :
Olomouc
DOI :
10.1109/ASRU.2013.6707768