Filters








14,305 Hits in 5.5 sec

On the feasibility of character n-grams pseudo-translation for Cross-Language Information Retrieval tasks

Jesús Vilares, Manuel Vilares, Miguel A. Alonso, Michael P. Oakes
2016 Computer Speech and Language  
The performance attained, close to the upper baseline, confirms the validity of character n-gram based approaches for Cross Language Information Retrieval tasks, both for indexing-retrieval and translation  ...  The field of Cross-Language Information Retrieval relates techniques close to both the Machine Translation and Information Retrieval fields, although in a context involving characteristics of its own.  ...  In response to this need, the aim of Cross-Language Information Retrieval (CLIR) is to provide techniques to return relevant documents written in a language (named the target language) different from the  ... 
doi:10.1016/j.csl.2015.09.004 fatcat:l3ctk3i7dzhw7h5tk54so54ggu

Dublin City University at CLEF 2005: Cross-Language Speech Retrieval (CL-SR) Experiments [chapter]

Adenike M. Lam-Adesina, Gareth J. F. Jones
2006 Lecture Notes in Computer Science  
We explored alternative translations methods with document retrieval based the combination of the multiple document fields using the BM25F field combination model.  ...  This approach has been widely used in cross-language information retrieval with good average performance when such an MT system is available for the language pair of the topic and document.  ...  Pseudo-Relevance Feedback Query expansion by pseudo relevance feedback (PRF) is a well-established procedure in both monolingual and cross-lingual IR, potentially providing some improvement in retrieval  ... 
doi:10.1007/11878773_87 fatcat:wp752kbvorg7hbvtudkasc2c6y

Dublin City University at CLEF 2006: Cross-Language Speech Retrieval (CL-SR) Experiments [chapter]

Gareth J. F. Jones, Ke Zhang, Adenike M. Lam-Adesina
2007 Lecture Notes in Computer Science  
We explored alternative translations methods with document retrieval based the combination of the multiple document fields using the BM25F field combination model.  ...  This approach has been widely used in cross-language information retrieval with good average performance when such an MT system is available for the language pair of the topic and document.  ...  Pseudo-Relevance Feedback Query expansion by pseudo relevance feedback (PRF) is a well-established procedure in both monolingual and cross-lingual IR, potentially providing some improvement in retrieval  ... 
doi:10.1007/978-3-540-74999-8_100 fatcat:icdb526fqjffbdzsg2gkylkdom

Cross-Language Access to Recorded Speech in the MALACH Project [chapter]

Douglas W. Oard, Dina Demner-Fushman, Jan Hajič, Bhuvana Ramabhadran, Samuel Gustman, William J. Byrne, Dagobert Soergel, Bonnie Dorr, Philip Resnik, Michael Picheny
2002 Lecture Notes in Computer Science  
It begins with a review of the state of the art in cross-language speech retrieval; approaches that will be investigated in the project are then described.  ...  Czech was selected as the first non-English language to be supported, so results of an initial experiment with Czech/English cross-language retrieval are reported.  ...  Cross-language text retrieval and monolingual speech retrieval are both wellresearched problems (see [10] for a survey of the first and [1] for a survey of the second), and these two challenges have  ... 
doi:10.1007/3-540-46154-x_8 fatcat:tdltd6g4lrclxigqputejmxk54

Efficient Speaker Recognition Using Approximated Cross Entropy (ACE)

Hagai Aronowitz, David Burshtein
2007 IEEE Transactions on Audio, Speech, and Language Processing  
In addition, a GMM compression algorithm is presented. This algorithm decreases considerably the storage needed for speaker retrieval.  ...  Gaussian mixture modeling is used for representing both training and test sessions and is shown to perform speaker recognition and retrieval extremely efficiently without any notable degradation in accuracy  ...  Speaker retrieval is an essential component of a speech retrieval system.  ... 
doi:10.1109/tasl.2007.902059 fatcat:h7l24x23avddreozfim57nzgyy

Identifying Effective Translations for Cross-lingual Arabic-to-English User-generated Speech Search

Ahmad Khwileh, Haithem Afli, Gareth Jones, Andy Way
2017 Proceedings of the Third Arabic Natural Language Processing Workshop  
Cross Language Information Retrieval (CLIR) systems are a valuable tool to enable speakers of one language to search for content of interest expressed in a different language.  ...  A key challenge in CLIR is crossing the language barrier between the query and the documents.  ...  We investigation concentrates on a cross-lingual user-generated speech search (CLUGS) task (Khwileh et al., 2015) .  ... 
doi:10.18653/v1/w17-1313 dblp:conf/wanlp/KhwilehAJW17 fatcat:yjpoqiez6bcq7kkl53v6zxscs4

An Adaptive Method for Cross-Recording Speaker Diarization

Gael Le Lan, Delphine Charlet, Anthony Larcher, Sylvain Meignier
2018 IEEE/ACM Transactions on Audio Speech and Language Processing  
Results indicate that the proposed adaptation framework reduces the cross-recording DER of 13% in average for variable collection sizes.  ...  Variability estimation is performed on consequent training datasets, which must be labeled by speaker.  ...  After the first iteration of adaptation, 198 seconds of true speech were retrieved, while 151 additional seconds were retrieved after the second iteration, for a total of 349 seconds.  ... 
doi:10.1109/taslp.2018.2844025 fatcat:evlmz4l73rdctkd43pna66yn3m

Cross-Modal Generalization Effects of Training Noncanonical Sentence Comprehension and Production in Agrammatic Aphasia

Beverly J. Jacobs, Cynthia K. Thompson
2000 Journal of Speech, Language and Hearing Research  
A single-subject multiple-baseline design across behaviors was used to measure acquisition and generalization within and across sentence types, as well as cross-modal generalization (i.e., from comprehension  ...  LST takes participants through a series of steps that emphasize the verb and verb argument structure, as well as the linguistic movement required to derive target sentences.  ...  A trace is left behind, marked with t, and a chain is established between the trace site and the moved NP.  ... 
doi:10.1044/jslhr.4301.05 pmid:10668649 pmcid:PMC3025283 fatcat:di3pm4jkxfhuhajzrmfproqbc4

Cross-Domain Adaptation of Spoken Language Identification for Related Languages: The Curious Case of Slavic Languages

Badr M. Abdullah, Tania Avgustinova, Bernd Möbius, Dietrich Klakow
2020 Interspeech 2020  
In this paper, we present a set of experiments to investigate the impact of domain mismatch on the performance of neural LID systems for a subset of six Slavic languages across two domains (read speech  ...  Our experiments show that (1) out-of-domain speech samples severely hinder the performance of neural LID models, and (2) while both spectral and cepstral features show comparable performance within-domain  ...  retrieval [5] .  ... 
doi:10.21437/interspeech.2020-2930 dblp:conf/interspeech/AbdullahAMK20 fatcat:f4kmope3qnhzfovn5qx65q7cly

Effects of cross-language voice training on speech perception: Whose familiar voices are more intelligible?

Susannah V. Levi, Stephen J. Winters, David B. Pisoni
2011 Journal of the Acoustical Society of America  
in which the talkers' voices were learned; knowledge of how a talker produces linguistically relevant contrasts in a particular language is necessary to increase speech intelligibility for words produced  ...  The absence of a processing advantage in speech intelligibility for the Germantrained listeners demonstrates limitations on the Familiar Talker Advantage, which crucially depends on the language context  ...  Research on memory encoding and retrieval processes reveals that performance in a test phase depends on the similarity between how information was initially encoded and how it is retrieved at test (Morris  ... 
doi:10.1121/1.3651816 pmid:22225059 pmcid:PMC3253604 fatcat:4qk5xd7c4rby3cmdgybn6w7egm

Cross-Linguistic Cognate Production in Spanish–English Bilingual Children With and Without Specific Language Impairment

Stephanie M. Grasso, Elizabeth D. Peña, Lisa M. Bedore, J. Gregory Hixon, Zenzi M. Griffin
2018 Journal of Speech, Language and Hearing Research  
In terms of speechlanguage intervention, exposure to a cognate in one language may increase vocabulary across both languages, which would grant cognates a baseline advantage in treatment.  ...  Basal and ceiling rules described by the manual were followed, and test administration in each language was discontinued if the child was unable to establish a basal.  ...  For any content that is copyright protected, please contact the copyright owner and seek written permission for publication by the American Speech-Language-Hearing Association (ASHA) in all forms (i.e.  ... 
doi:10.1044/2017_jslhr-l-16-0421 pmid:29466535 pmcid:PMC5963043 fatcat:qhujwupwo5e6tnyobfuimkn6yy

Efficacy and Cross-Domain Effects of a Morphosyntax and a Phonology Intervention

Ann A. Tyler, Kerry E. Lewis, Allison Haskill, Leslie C. Tolbert
2002 Language, Speech & Hearing Services in Schools  
The morphosyntax first sequence also led to slightly better overall morphosyntactic performance.  ...  The morphosyntax intervention led to cross-domain change in phonology that was similar to that achieved by the phonology intervention.  ...  We also wish to express our thanks to the children, their parents, and the Washoe County School District speech-language pathologists and teachers who participated in this research.  ... 
doi:10.1044/0161-1461(2002/005) pmid:27764416 fatcat:2aizfx5xjveobnaq255iyn2api

Exploiting redundancy in cross-channel video retrieval

Bouke Huurnink, Maarten de Rijke
2007 Proceedings of the international workshop on Workshop on multimedia information retrieval - MIR '07  
We describe this phenomenon, and use it to develop a framework to incorporate redundancy for cross-channel retrieval of visual items using speech.  ...  Testing our models in a series of retrieval experiments, we find that incorporating the fact that information occurs redundantly into cross-channel retrieval leads to significant improvements in retrieval  ...  As our baseline we use a search using shot speech only (i.e., not expanded to include speech from neighbouring shots).  ... 
doi:10.1145/1290082.1290109 dblp:conf/mir/HuurninkR07 fatcat:mmz7ud4eyzcb7ki34mdnlzqstm

The TREC Spoken Document Retrieval Track

Ellen Voorhees, John Garofolo
2005 Bulletin of the American Society for Information Science and Technology  
The Text REtrieval Conference TREC workshops provide a forum for di erent groups to compare retrieval systems on common retrieval tasks.  ...  The 1997 TREC workshop will feature a Spoken Document Retrieval task for the rst time.  ...  For SDR Group participants only Baseline B1: a retrieval run using the the Baseline Speech Recognizer Transcriptions as input. This is the speech run" for QSDR participants.  ... 
doi:10.1002/bult.170 fatcat:rbf2z3tevzb57gxrdxqjkrhj4q

Multi-modal Transformer for Video Retrieval [article]

Valentin Gabeur, Chen Sun, Karteek Alahari, Cordelia Schmid
2020 arXiv   pre-print
The task of retrieving video content relevant to natural language queries plays a critical role in effectively handling internet-scale datasets.  ...  Most of the existing methods for this caption-to-video retrieval problem do not fully exploit cross-modal cues present in video.  ...  We incorporate this video encoder along with a caption encoder in a cross-modal framework to perform Table 4 : Retrieval performance on the MSRVTT dataset. 1k-A and 1k-B denote test sets of 1000 randomly  ... 
arXiv:2007.10639v1 fatcat:r2duzkctwvbffetignqkmxjoka
« Previous Showing results 1 — 15 out of 14,305 results