Spoken term detection using fast phonetic decoding

Wallace, Roy; Vogt, Robbie; Sridharan, Sridha

doi:10.1109/icassp.2009.4960725

Cited by 15 publications

(13 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We call this lexical deviation, which is quite different from acoustic variation and therefore can not be fully compensated for by commonly employed soft match techniques (e.g. [21]- [24]). …”

Section: Motivationsmentioning

confidence: 99%

“…A widely used approach to compute P (Q|Q) is based on a confusion matrix [21]- [24], [47]- [50]. In this approach, the insertion/deletion/substitution probabilities of phoneme pairs are estimated by a forced alignment between phoneme recognition output on the development set and the canonical transcription, which forms a confusion matrix that represents the match degree of a phoneme pair (a special null phoneme is included to allow insertions and deletions).…”

Section: Soft Matchmentioning

confidence: 99%

“…Better LTS models can be used, such as joint-multigram models [6]. Soft match is the most common technique for mitigating acoustic variation; it allows for some mismatch between the pronunciation predicted for the search term and the phoneme sequences in the lattice and typically involves a penalty based on either edit distance [13], [36], [37], acoustic confusion [21], [22], [24], [38] or model distance [39], [40]. Lexical deviation, however, has not been widely investigated until recently [3], [25].…”

Section: A Oov Uncertaintymentioning

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

Section: Motivationsmentioning

confidence: 99%

Section: Soft Matchmentioning

confidence: 99%

Section: A Oov Uncertaintymentioning

confidence: 99%

See 1 more Smart Citation

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

show abstract

“…By doing so, the OOV problem is mostly solved since subword dictionaries have much higher coverage than word-based dictionaries. Some typical subword units are phones [56][57][58][59][60][61], syllables [62][63][64], morpheme [65].…”

Section: Challenges and Existing Approaches For Stdmentioning

confidence: 99%

“…Typically, phonemes are the common linguistical subword units to handle OOV queries. The phoneme-based approach was first proposed for SDR task in [68,92,120] and has been applied for STD in [56][57][58][59][60][61]. In this approach, a query is converted to a phoneme sequences using a pronunciation dictionary or a Letter-to-Sound tool for OOV queries.…”

Section: Linguistical Subword Unitsmentioning

confidence: 99%