Vocabulary independent spoken term detection

Mamou, Jonathan; Ramabhadran, Bhuvana; Siohan, Olivier

doi:10.1145/1277741.1277847

Cited by 162 publications

(99 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, as noted by Logan et al (2000), about 12% of users' queries typically contain out-of-vocabulary (OOV) words, which will never be found in the word lattices, because they do not appear in the LVCSR system vocabulary. Common approaches to solve this problem usually involve producing sub-word (typically phone/phoneme) lattices with the ASR subsystem, and then searching for sub-word representations of the enquiry terms (Saraçlar and Sproat, 2004;Mamou et al, 2007;Can et al, 2009;Szöke et al, 2006;Wallace et al, 2007;Parlak and Saraçlar, 2008). Other sub-word units are possible, such as syllables (Meng et al, 2007), graphemes Tejedor et al, 2008) or multi-grams (Pinto et al, 2008;Szöke et al, 2008a).…”

Section: Spoken Term Detectionmentioning

confidence: 99%

Feature analysis for discriminative confidence estimation in spoken term detection

Tejedor

Toledano

Wang

et al. 2014

Computer Speech & Language

View full text Add to dashboard Cite

El acceso a la versión del editor puede requerir la suscripción del recurso Access to the published version may require subscription AbstractDiscriminative confidence based on multi-layer perceptrons (MLPs) and multiple features has shown significant advantage compared to the widely used lattice-based confidence in spoken term detection (STD). Although the MLP-based framework can handle any features derived from a multitude of sources, choosing all possible features may lead to over complex models and hence less generality. In this paper, we design an extensive set of features and analyze their contribution to STD individually and as a group. The main goal is to choose a small set of features that are sufficiently informative while keeping the model simple and generalizable. We employ two established models to conduct the analysis: one is linear regression which targets for the most relevant features and the other is logistic linear regression which targets for the most discriminative features. We find the most informative features are comprised of those derived from diverse sources (ASR decoding, duration and lexical properties) and the two models deliver highly consistent feature ranks. STD experiments on both English and Spanish data demonstrate significant performance gains with the proposed feature sets.

show abstract

Section: Spoken Term Detectionmentioning

confidence: 99%

Feature analysis for discriminative confidence estimation in spoken term detection

Tejedor

Toledano

Wang

et al. 2014

Computer Speech & Language

View full text Add to dashboard Cite

show abstract

“…The usual approach to detecting OOV terms employs subword units [6], [8], [20]: search terms are converted to a subword sequence (usually phonemes) by letter-to-sound (LTS) conversion. This sequence is then searched for in previouslygenerated subword lattices or transcripts [11], [20].…”

Section: Motivationsmentioning

confidence: 99%

“…This sequence is then searched for in previouslygenerated subword lattices or transcripts [11], [20]. In this paper, we use a phoneme-based system.…”

Section: Motivationsmentioning

confidence: 99%

“…With SPM, however, the bias problem is more significant. On one hand, more assumptions have been introduced; on the other hand, the interpolation factor inevitably changes the value of the confidence (see (14)). The highly biased confidence makes optimising the linear remedy rather challenging, especially when this optimisation is interweaved with the optimisation of the interpolation factor.…”

Section: B Spm and Confidence Biasmentioning

confidence: 99%

“…where Q has replaced K as it is determinately generated by K, and a layer separation assumption has again been applied to get from (19) to (20). Similarly to SPM, this indicates that detecting a term (or its pronunciation) is equal to detecting all the pronunciations allowed by soft match, and the confidence of detecting the term can be computed by summing the confidences of all detections being found with these 'soft matched' pronunciations sharing the same starting and ending time, if we define the confidence of a single detection as:…”

Section: Soft Matchmentioning

confidence: 99%

See 2 more Smart Citations

Stochastic Pronunciation Modeling for Out-of-Vocabulary Spoken Term Detection

Wang

King

Frankel

2011

IEEE Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Abstract-Spoken term detection (STD) is the name given to the task of searching large amounts of audio for occurrences of spoken terms, which are typically single words or short phrases. One reason that STD is a hard task is that search terms tend to contain a disproportionate number of out-of-vocabulary (OOV) words. The most common approach to STD uses subword units. This, in conjunction with some method for predicting pronunciations of OOVs from their written form, enables the detection of OOV terms but performance is considerably worse than for in-vocabulary terms. This performance differential can be largely attributed to the special properties of OOVs.One such property is the high degree of uncertainty in the pronunciation of OOVs. We present a stochastic pronunciation model (SPM) which explicitly deals with this uncertainty. The key insight is to search for all possible pronunciations when detecting an OOV term, explicitly capturing the uncertainty in pronunciation. This requires a probabilistic model of pronunciation, able to estimate a distribution over all possible pronunciations. We use a joint-multigram model (JMM) for this and compare the JMM-based SPM with the conventional soft match approach. Experiments using speech from the meetings domain demonstrate that the SPM performs better than soft match in most operating regions, especially at low false alarm probabilities. Furthermore, SPM and soft match are found to be complementary: their combination provides further performance gains.

show abstract

Speech Retrieval

Chelba¹,

Hazen²,

Ramabhadran³

et al. 2011

Spoken Language Understanding

Self Cite

View full text Add to dashboard Cite

In this chapter we discuss the retrieval and browsing of spoken audio documents. We focus primarily on the application of document search where a user provides a query and the system returns a set of audio documents that best match the query. The primary technical challenges of speech retrieval lie in the retrieval system's ability to deal with imperfect speech recognition technology that produces errorful output due to misrecognitions cause by inadequate statistical models or out-of-vocabulary words. This chapter provides an overview of the common tasks and data sets, evaluation metrics, and algorithms most commonly used in this growing area of research.

show abstract

Vocabulary independent spoken term detection

Cited by 162 publications

References 15 publications

Feature analysis for discriminative confidence estimation in spoken term detection

Feature analysis for discriminative confidence estimation in spoken term detection

Stochastic Pronunciation Modeling for Out-of-Vocabulary Spoken Term Detection

Speech Retrieval

Contact Info

Product

Resources

About