Out-of-vocabulary word detection in a speech-to-speech translation system

Kuo, Hong-Kwang Jeff; Kislal, Ellen Eide; Mangu, Lidia; Soltau, Hagen; Beran, Tomáš

doi:10.1109/icassp.2014.6854979

Cited by 6 publications

(2 citation statements)

References 14 publications

(15 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Earlier work often used a combination of one-best word and phone sequences [6]. More recent work uses word confusion networks to represent the ASR hypotheses [7,8]. These approaches are combined in [9].…”

Section: Introductionmentioning

confidence: 99%

Effective data-driven feature learning for detecting name errors in automatic speech recognition

Marin

Ostendorf

2014

2014 IEEE Spoken Language Technology Workshop (SLT)

View full text Add to dashboard Cite

This paper addresses the problem of detecting name errors in automatic speech recognition (ASR) output. The highly skewed label distributions (i.e. name errors are infrequent), sparse training data, and large number of potential lexical features pose significant challenges for training name error classification systems. Data-driven feature learning is needed for handling multiple languages but is sensitive to over fitting. We address the problem by designing aggregate features using a related (sentence-level name detection) task, and reduce dimensionality of the lexical features using word classes. Experiments on conversational domain data in both English and Iraqi Arabic show that best results are obtained using all feature mapping methods plus feature selection using L1 regularization.

show abstract

Section: Introductionmentioning

confidence: 99%

Effective data-driven feature learning for detecting name errors in automatic speech recognition

Marin

Ostendorf

2014

2014 IEEE Spoken Language Technology Workshop (SLT)

View full text Add to dashboard Cite

show abstract

“…OOV word detection has been studied over the years, and methods based on the word/fragment hybrid ASR approach are widely used [1,2,3,4,5]. Hybrid ASR uses a hybrid lexicon consisting of not only words but also subword sequences (fragments) and a hybrid language model (LM) trained on texts in which low frequency words are replaced by fragment sequences.…”

Section: Introductionmentioning

confidence: 99%

Recurrent Out-of-Vocabulary Word Detection Using Distribution of Features

et al. 2016

View full text Add to dashboard Cite

The repeated use of out-of-vocabulary (OOV) words in a spoken document seriously degrades a speech recognizer's performance. This paper provides a novel method for accurately detecting such recurrent OOV words. Standard OOV word detection methods classify each word segment into in-vocabulary (IV) or OOV. This word-by-word classification tends to be affected by sudden vocal irregularities in spontaneous speech, triggering false alarms. To avoid this sensitivity to the irregularities, our proposal focuses on consistency of the repeated occurrence of OOV words. The proposed method preliminarily detects recurrent segments, segments that contain the same word, in a spoken document by open vocabulary spoken term discovery using a phoneme recognizer. If the recurrent segments are OOV words, features for OOV detection in those segments should exhibit consistency. We capture this consistency by using the mean and variance (distribution) of features (DOF) derived from the recurrent segments, and use the DOF for IV/OOV classification. Experiments illustrate that the proposed method's use of the DOF significantly improves its performance in recurrent OOV word detection.

show abstract