The spoken web search task at MediaEval 2012

Metze, Florian; Anguera, Xavier; Barnard, Etienne; Davel, Marelie H.; Gravier, Guillaume

doi:10.1109/icassp.2013.6639247

Cited by 41 publications

(33 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As an alternative to the widespread template matching approaches, other systems use Acoustic Keyword Spotting (AKWS) [12,13], exploiting in several ways acoustic models in multiple languages. A review of these and other methods can be found in [3,4].…”

Section: Query-by-example Speech Search (Qess)mentioning

confidence: 99%

“…In this "query-by-example" scenario, instances of these key terms in the test data to be mined are discovered through their acoustic match to the provided spoken examples. Query-byExample Speech Search (QESS) is a particularly relevant problem for low-resource languages, and has recently gained significant research interest partially due to the success of the Spoken Web Search (SWS) task at the MediaEval evaluation series [3,4]. In practice, the query-by-example task can be considered as a sort of generalization of the problem of speech search based on text queries.…”

Section: Query-by-example Speech Search (Qess)mentioning

confidence: 99%

See 1 more Smart Citation

Privacy-preserving Query-by-Example Speech Search

Portêlo

Abad

Raj

et al. 2015

2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

This paper investigates a new privacy-preserving paradigm for the task of Query-by-Example Speech Search using Secure Binary Embeddings, a hashing method that converts vector data to bit strings through a combination of random projections followed by banded quantization. The proposed method allows performing spoken query search in an encrypted domain, by analyzing ciphered information computed from the original recordings. Unlike other hashing techniques, the embeddings allow the computation of the distance between vectors that are close enough, but are not perfect matches. This paper shows how these hashes can be combined with Dynamic Time Warping based on posterior derived features to perform secure speech search. Experiments performed on a sub-set of the SpeechDat Portuguese corpus showed that the proposed privacy-preserving system obtains similar results to its non-private counterpart.

show abstract

Section: Query-by-example Speech Search (Qess)mentioning

confidence: 99%

Section: Query-by-example Speech Search (Qess)mentioning

confidence: 99%

Privacy-preserving Query-by-Example Speech Search

Portêlo

Abad

Raj

et al. 2015

2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…Most of the existing QbyE methods [4], [6], [7], [8], [9], [10] and STD approaches [11], [12], [13], [14] use DTW and its variations [3], [15], [16]. Over the past decade, mass of the related research is focused on novelty of the template representation methods [17], [18], [19], [20], [21].…”

Section: Introductionmentioning

confidence: 99%

“…Over the past decade, mass of the related research is focused on novelty of the template representation methods [17], [18], [19], [20], [21]. An acoustic segmentation model based STD is presented in [10] that amalgamates the self-organising models, query matching, and query modelling processes.…”

Section: Introductionmentioning

confidence: 99%

An Intelligent System for Spoken Term Detection That Uses Belief Combination

Khan¹,

Kuru²

2017

IEEE Intell. Syst.

View full text Add to dashboard Cite

Spoken Term Detection (STD) can be considered as a sub-part of the automatic speech recognition which aims to extract the partial information from speech signals in the form of query utterances. A variety of STD techniques available in the literature employ a single source of evidence for the query utterance match/mismatch determination. In this manuscript, we develop an acoustic signal processing based approach for STD that incorporates a number of techniques for silence removal, dynamic noise filtration, and evidence combination using Dempster-Shafer Theory (DST). A "spectral-temporal features based voiced segment detection" and "energy and zero cross rate based unvoiced segment detection" are built to remove the silence segments in the speech signal. Comprehensive experiments have been performed on large speech datasets and consequently satisfactory results have been achieved with the proposed approach. Our approach improves the existing speaker dependent STD approaches, specifically the reliability of query utterance spotting by combining the evidences from multiple belief sources.Keywords: Spoken term detection, Acoustic keyword spotting, Query-by-example, Dempster-Shafer"s theory, Speech recognition, Speech processing. Acknowledgement:A special gratitude we give to Prof. Daniel Neagu, University of Bradford, whose contribution in stimulating suggestions helped us to coordinate this research in terms of statistical analysis, and performance evaluation methods.

show abstract

“…Recently, KWS under limited-resource conditions [19][20][21][22][23] has become a research focus because training data collection is often one of the most time-consuming and expensive efforts in the overall system building process. While there are more than thousands of languages in the world [24] recorded in many different conditions, it is usually not practical for KWS system designers to collect and transcribe a great amount of training speech data for every language of interest in a particular environment.…”

Section: Introductionmentioning

confidence: 99%

A Keyword-Aware Language Modeling Approach to Spoken Keyword Search

Chen

Lim

et al. 2015

J Sign Process Syst

View full text Add to dashboard Cite

A keyword-sensitive language modeling framework for spoken keyword search (KWS) is proposed to combine the advantages of conventional keyword-filler based and large vocabulary continuous speech recognition (LVCSR) based KWS systems. The proposed framework allows keyword search systems to be flexible on keyword target settings as in the LVCSR-based keyword search. In low-resource scenarios it facilitates KWS with an ability to achieve high keyword detection accuracy as in the keyword-filler based systems and to attain a low false alarm rate inherent in the LVCSRbased systems. The proposed keyword-aware grammar is realized by incorporating keyword information to re-train and modify the language models used in LVCSR-based KWS. Experimental results, on the evalpart1 data of the IARPA Babel OpenKWS13 Vietnamese tasks, indicate that the proposed approach achieves a relative improvement, over the conventional LVCSR-based KWS systems, of the actual term weighted value for about 57 % (from 0.2093 to 0.3287) and 20 % (from 0.4578 to 0.5486) on the limited-language-pack and full-language-pack tasks, respectively.

show abstract

The spoken web search task at MediaEval 2012

Cited by 41 publications

References 15 publications

Privacy-preserving Query-by-Example Speech Search

Privacy-preserving Query-by-Example Speech Search

An Intelligent System for Spoken Term Detection That Uses Belief Combination

A Keyword-Aware Language Modeling Approach to Spoken Keyword Search

Contact Info

Product

Resources

About