Reshaping automatic speech transcripts for robust high-level spoken document analysis

Fayolle, J; Moreau, Fabienne; Raymond, Christian; Gravier, Guillaume

doi:10.1145/1871840.1871849

Cited by 5 publications

(6 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We will study the bias of the posterior probabilities, as it was shown in [21] to impact the NCE and thus the quality of confidence measures based on these probabilities. We will investigate using other features for the characterization step and/or the detection step, especially probabilities given by n-class LMs using POS classes (as the n-class part of our word-to-class backoff LM) that were proven to be efficient, in recent works in speech recognition [13,36]. It would also be of interest to investigate using other measures to optimize the error detection threshold, like the F-measure, and to study how it impacts the correction step and thus the overall error handling process.…”

Section: Resultsmentioning

confidence: 99%

“…They are also used in machine translation tasks to assess the quality of the translation, as in [40], where various measures are compared. Among all these confidence measures, word posterior probabilities have been shown to be among the best [5,23,41], and they can be combined with other information sources in a neural network [23], in a SVM [21], or using conditional random fields [13] to achieve even better results. These confidence measures can also be used to detect recognition errors by rejecting words, the value of which is below a considered threshold.…”

Section: Speech Recognitionmentioning

confidence: 99%

“…To evaluate the performance of the task, from a recognition point of view, we use the common word accuracy (WA) and word recognition rate (WRR), which are defined by WA = #wor ds − #subst − #del − #ins #wor ds (13) and WRR = #wor ds − #subst − #del #wor ds (14) with #wor ds being the number of all the words of the phrases to be recognized in the set considered and #subst, #ins, and #del the number of substitutions, insertions, and deletions in the resulting recognized phrases, respectively. To evaluate the posterior probabilities used as confidence indices, we compute the normalized cross entropy (NCE), which is commonly used to measure the quality of confidence measures (the higher its value, the better the confidence measure).…”

Section: Evaluation Metricsmentioning

confidence: 99%

“…Techniques from natural language processing applications thus need to be adapted to deal with such noisy data. In the field of noisy data processing, works have been focusing more particularly on dealing with text documents [39], such as blogs…, or transcripts from automatic speech recognition (ASR) systems [13,36]. However, there have been few works that deal with transcripts from handwriting recognition (HWR) systems [26,35].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Error handling approach using characterization and correction steps for handwritten document analysis

Quiniou

Cheriet

Anquetil

2011

IJDAR

View full text Add to dashboard Cite

In this paper, we present a framework to handle recognition errors from a N -best list of output phrases given by a handwriting recognition system, with the aim to use the resulting phrases as inputs to a higher-level application. The framework can be decomposed into four main steps: phrase alignment, detection, characterization, and correction of word error hypotheses. First, the N -best phrases are aligned to the top-list phrase, and word posterior probabilities are computed and used as confidence indices to detect word error hypotheses on this top-list phrase (in comparison with a learned threshold). Then, the errors are characterized into predefined types, using the word posterior probabilities of the top-list phrase and other features to feed a trained SVM. Finally, the final output phrase is retrieved, thanks to a correction step that used the characterized error hypotheses and a designed word-to-class backoff language model. First experiments were conducted on the ImadocSen-OnDB handwritten sentence database and on the IAM-OnDB handwritten text database, using two recognizers. We present first results on an implementation of the proposed framework for handling recognition errors on transcripts of handwritten phrases provided by recognition systems.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Speech Recognitionmentioning

confidence: 99%

Section: Evaluation Metricsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Error handling approach using characterization and correction steps for handwritten document analysis

Quiniou

Cheriet

Anquetil

2011

IJDAR

View full text Add to dashboard Cite

show abstract

“…Accessing large collections of audio files is difficult and intrinsically time-consuming. Browsing audio recordings is a difficult task for humans, and reviewing such documents quickly is technically difficult [Fayolle et al, 2010]. Automatic speech recognition (ASR) produces searchable text but is also expensive in terms of both time and computational resources.…”

Section: Motivationmentioning

confidence: 99%

Factoid question answering for spoken documents

Comas Umbert

View full text Add to dashboard Cite

In this dissertation, we present a factoid question answering system, specifically tailored for Question Answering (QA) on spoken documents. This work explores, for the first time, which techniques can be robustly adapted from the usual QA on written documents to the more difficult spoken documents scenario. More specifically, we study new information retrieval (IR) techniques designed for speech, and utilize several levels of linguistic information for the speech-based QA task. These include named-entity detection with phonetic information, syntactic parsing applied to speech transcripts, and the use of coreference resolution. Our approach is largely based on supervised machine learning techniques, with special focus on the answer extraction step, and makes little use of handcrafted knowledge. Consequently, it should be easily adaptable to other domains and languages. In the work resulting of this Thesis, we have impulsed and coordinated the creation of an evaluation framework for the task of QA on spoken documents. The framework, named QAst, provides multi-lingual corpora, evaluation questions, and answers key. These corpora have been used in the QAst evaluation that was held in the CLEF workshop for the years 2007, 2008 and 2009, thus helping the developing of state-of-the-art techniques for this particular topic. The presentend QA system and all its modules are extensively evaluated on the European Parliament Plenary Sessions English corpus composed of manual transcripts and automatic transcripts obtained by three different Automatic Speech Recognition (ASR) systems that exhibit significantly different word error rates. This data belongs to the CLEF 2009 track for QA on speech transcripts. The main results confirm that syntactic information is very useful for learning to rank question candidates, improving results on both manual and automatic transcripts unless the ASR quality is very low. Overall, the performance of our system is comparable or better than the state-of-the-art on this corpus, confirming the validity of our approach. En aquesta Tesi, presentem un sistema de Question Answering (QA) factual, especialment ajustat per treballar amb documents orals. En el desenvolupament explorem, per primera vegada, quines tècniques de les habitualment emprades en QA per documents escrit són suficientment robustes per funcionar en l'escenari més difícil de documents orals. Amb més especificitat, estudiem nous mètodes de Information Retrieval (IR) dissenyats per tractar amb la veu, i utilitzem diversos nivells d'informació linqüística. Entre aquests s'inclouen, a saber: detecció de Named Entities utilitzant informació fonètica, "parsing" sintàctic aplicat a transcripcions de veu, i també l'ús d'un sub-sistema de detecció i resolució de la correferència. La nostra aproximació al problema es recolza en gran part en tècniques supervisades de Machine Learning, estant aquestes enfocades especialment cap a la part d'extracció de la resposta, i fa servir la menor quantitat possible de coneixement creat per humans. En conseqüència, tot el procés de QA pot ser adaptat a altres dominis o altres llengües amb relativa facilitat. Un dels resultats addicionals de la feina darrere d'aquesta Tesis ha estat que hem impulsat i coordinat la creació d'un marc d'avaluació de la taska de QA en documents orals. Aquest marc de treball, anomenat QAst (Question Answering on Speech Transcripts), proporciona un corpus de documents orals multi-lingüe, uns conjunts de preguntes d'avaluació, i les respostes correctes d'aquestes. Aquestes dades han estat utilitzades en les evaluacionis QAst que han tingut lloc en el si de les conferències CLEF en els anys 2007, 2008 i 2009; d'aquesta manera s'ha promogut i ajudat a la creació d'un estat-de-l'art de tècniques adreçades a aquest problema en particular. El sistema de QA que presentem i tots els seus particulars sumbòduls, han estat avaluats extensivament utilitzant el corpus EPPS (transcripcions de les Sessions Plenaries del Parlament Europeu) en anglès, que cónté transcripcions manuals de tots els discursos i també transcripcions automàtiques obtingudes mitjançant tres reconeixedors automàtics de la parla (ASR) diferents. Els reconeixedors tenen característiques i resultats diferents que permetes una avaluació quantitativa i qualitativa de la tasca. Aquestes dades pertanyen a l'avaluació QAst del 2009. Els resultats principals de la nostra feina confirmen que la informació sintàctica és mol útil per aprendre automàticament a valorar la plausibilitat de les respostes candidates, millorant els resultats previs tan en transcripcions manuals com transcripcions automàtiques, descomptat que la qualitat de l'ASR sigui molt baixa. En general, el rendiment del nostre sistema és comparable o millor que els altres sistemes pertanyents a l'estat-del'art, confirmant així la validesa de la nostra aproximació.

show abstract

Recapitulization of tweets using graph-based clustering

Lobo¹,

Ansari²

2017

2017 2nd International Conference on Communication Systems, Computing and IT Applications (CSCITA)

View full text Add to dashboard Cite

Reshaping automatic speech transcripts for robust high-level spoken document analysis

Cited by 5 publications

References 14 publications

Error handling approach using characterization and correction steps for handwritten document analysis

Error handling approach using characterization and correction steps for handwritten document analysis

Factoid question answering for spoken documents

Recapitulization of tweets using graph-based clustering

Contact Info

Product

Resources

About