Pub Date : 2011-12-01DOI: 10.1109/ASRU.2011.6163943
L. Mangu, H. Kuo, Stephen M. Chu, Brian Kingsbury, G. Saon, H. Soltau, Fadi Biadsy
We describe the Arabic broadcast transcription system fielded by IBM in the GALE Phase 4 machine translation evaluation. Key advances over our Phase 3.5 system include improvements to context-dependent modeling in vowelized Arabic acoustic models; the use of neural-network features provided by the International Computer Science Institute; Model M language models; a neural network language model that uses syntactic and morphological features; and improvements to our system combination strategy. These advances were instrumental in achieving a word error rate of 8.9% on the Phase 4 evaluation set, and an absolute improvement of 1.6% word error rate over our 2008 system on the unsequestered Phase 3.5 evaluation data.
{"title":"The IBM 2009 GALE Arabic speech transcription system","authors":"L. Mangu, H. Kuo, Stephen M. Chu, Brian Kingsbury, G. Saon, H. Soltau, Fadi Biadsy","doi":"10.1109/ASRU.2011.6163943","DOIUrl":"https://doi.org/10.1109/ASRU.2011.6163943","url":null,"abstract":"We describe the Arabic broadcast transcription system fielded by IBM in the GALE Phase 4 machine translation evaluation. Key advances over our Phase 3.5 system include improvements to context-dependent modeling in vowelized Arabic acoustic models; the use of neural-network features provided by the International Computer Science Institute; Model M language models; a neural network language model that uses syntactic and morphological features; and improvements to our system combination strategy. These advances were instrumental in achieving a word error rate of 8.9% on the Phase 4 evaluation set, and an absolute improvement of 1.6% word error rate over our 2008 system on the unsequestered Phase 3.5 evaluation data.","PeriodicalId":254007,"journal":{"name":"2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114618550","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2011-07-12DOI: 10.1109/ICASSP.2011.5947264
D. Schmid, G. Enzner
The blind identification of single-input multiple-output (SIMO) systems suffers in the presence of near-common and exact common zeros between the channels, particularly in conjunction with observation noise. In general, we notice an ambiguity of the identification which cannot be resolved without further a priori information on the channel coefficients. In order to enable an adequate evaluation of blind SIMO identification in such cases, we develop the normalized filter-projection misalignment (NFPM), which represents a multichannel squared-error distance between true and estimated channels, while absorbing a common filter error due to a possible lack of identifiability. Using the NFPM measure, we demonstrate experimentally that the steady-state performance of the blind multichannel least mean-square (MCLMS) algorithm in the presence of missing channel diversity and noise is in line with the results obtained from supervised least mean-square (LMS) system identification.
{"title":"Evaluation of adaptive blind SIMO identification in terms of a normalized filter-projection misalignment","authors":"D. Schmid, G. Enzner","doi":"10.1109/ICASSP.2011.5947264","DOIUrl":"https://doi.org/10.1109/ICASSP.2011.5947264","url":null,"abstract":"The blind identification of single-input multiple-output (SIMO) systems suffers in the presence of near-common and exact common zeros between the channels, particularly in conjunction with observation noise. In general, we notice an ambiguity of the identification which cannot be resolved without further a priori information on the channel coefficients. In order to enable an adequate evaluation of blind SIMO identification in such cases, we develop the normalized filter-projection misalignment (NFPM), which represents a multichannel squared-error distance between true and estimated channels, while absorbing a common filter error due to a possible lack of identifiability. Using the NFPM measure, we demonstrate experimentally that the steady-state performance of the blind multichannel least mean-square (MCLMS) algorithm in the presence of missing channel diversity and noise is in line with the results obtained from supervised least mean-square (LMS) system identification.","PeriodicalId":254007,"journal":{"name":"2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128038259","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2011-07-12DOI: 10.1109/ICASSP.2011.5947561
Sunghwan Shin, H. Jung, B. Juang
In this paper, we follow the minimum error principle for acoustic modeling and formulate error objectives in insertion, deletion, and substitution separately for minimization during training. This new training paradigm generalized from the MVE criterion can explain the direct relationship between recognition errors and detection errors by re-interpreting deletion, insertion, and substitution errors as miss, false alarm, and miss/false-alarm errors happening together. Under the MVE criterion, by applying two mis-verification measures for miss and false alarm errors selectively along with the types of recognition error definition, we developed three individual objective training criteria, minimum deletion error (MDE), minimum insertion error (MIE), and minimum substitution error (MSE), of which each objective function can directly minimize each of the three types of the recognition errors. In the TIMIT phone recognition task, the experimental results confirm that each objective criterion of MDE, MIE, and MSE results in primarily minimizing its target error type, respectively. Furthermore, a simple combination of the individual objective criteria outperforms the conventional string-based MCE in the overall recognition error rate.
{"title":"Discriminative Training for direct minimization of deletion, insertion and substitution errors","authors":"Sunghwan Shin, H. Jung, B. Juang","doi":"10.1109/ICASSP.2011.5947561","DOIUrl":"https://doi.org/10.1109/ICASSP.2011.5947561","url":null,"abstract":"In this paper, we follow the minimum error principle for acoustic modeling and formulate error objectives in insertion, deletion, and substitution separately for minimization during training. This new training paradigm generalized from the MVE criterion can explain the direct relationship between recognition errors and detection errors by re-interpreting deletion, insertion, and substitution errors as miss, false alarm, and miss/false-alarm errors happening together. Under the MVE criterion, by applying two mis-verification measures for miss and false alarm errors selectively along with the types of recognition error definition, we developed three individual objective training criteria, minimum deletion error (MDE), minimum insertion error (MIE), and minimum substitution error (MSE), of which each objective function can directly minimize each of the three types of the recognition errors. In the TIMIT phone recognition task, the experimental results confirm that each objective criterion of MDE, MIE, and MSE results in primarily minimizing its target error type, respectively. Furthermore, a simple combination of the individual objective criteria outperforms the conventional string-based MCE in the overall recognition error rate.","PeriodicalId":254007,"journal":{"name":"2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129985230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2011-07-12DOI: 10.1109/ICASSP.2011.5947721
Andrew Harms, W. Bajwa, A. Calderbank
Technological constraints severely limit the rate at which analog-to-digital converters can reliably sample signals. Recently, Tropp et al. proposed an architecture, termed the random demodulator (RD), that attempts to overcome this obstacle for sparse bandlimited signals. One integral component of the RD architecture is a white noise-like, bipolar modulating waveform that changes polarity at a rate equal to the signal bandwidth. Since there is a hardware limitation to how fast analog waveforms can change polarity without undergoing shape distortion, this leads to the RD also having a constraint on the maximum allowable bandwidth. In this paper, an extension of the RD, termed the constrained random demodulator (CRD), is proposed that bypasses this bottleneck by replacing the original modulating waveform with a run-length limited (RLL) modulating waveform that changes polarity at a slower rate than the signal bandwidth. One of the main contributions of the paper is establishing that the CRD, despite employing a modulating waveform with correlations, enjoys some theoretical guarantees for certain RLL waveforms. In addition, for a given sampling rate and rate of change in the modulating waveform polarity, numerical simulations confirm that the CRD, using an appropriate RLL waveform, can sample a signal with an even wider bandwidth without a significant loss in performance.
{"title":"Beating nyquist through correlations: A constrained random demodulator for sampling of sparse bandlimited signals","authors":"Andrew Harms, W. Bajwa, A. Calderbank","doi":"10.1109/ICASSP.2011.5947721","DOIUrl":"https://doi.org/10.1109/ICASSP.2011.5947721","url":null,"abstract":"Technological constraints severely limit the rate at which analog-to-digital converters can reliably sample signals. Recently, Tropp et al. proposed an architecture, termed the random demodulator (RD), that attempts to overcome this obstacle for sparse bandlimited signals. One integral component of the RD architecture is a white noise-like, bipolar modulating waveform that changes polarity at a rate equal to the signal bandwidth. Since there is a hardware limitation to how fast analog waveforms can change polarity without undergoing shape distortion, this leads to the RD also having a constraint on the maximum allowable bandwidth. In this paper, an extension of the RD, termed the constrained random demodulator (CRD), is proposed that bypasses this bottleneck by replacing the original modulating waveform with a run-length limited (RLL) modulating waveform that changes polarity at a slower rate than the signal bandwidth. One of the main contributions of the paper is establishing that the CRD, despite employing a modulating waveform with correlations, enjoys some theoretical guarantees for certain RLL waveforms. In addition, for a given sampling rate and rate of change in the modulating waveform polarity, numerical simulations confirm that the CRD, using an appropriate RLL waveform, can sample a signal with an even wider bandwidth without a significant loss in performance.","PeriodicalId":254007,"journal":{"name":"2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130542614","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2011-07-12DOI: 10.1109/ICASSP.2011.5947461
Line Adde, T. Svendsen
In this paper, the task of selecting the optimal subset of pronunciation variants from a set of automatically generated candidates is recast as a tree search problem. In this approach, the optimal recognition lexicon corresponds with the optimal path through a search tree. We define a discriminative evaluation function to guide the search algorithm, which is based on estimates of the number of recognition errors before and after a lexicon change. The error rate for a given lexicon is estimated using the Minimum Classification Error framework. Selecting pronunciation candidates by means of this search algorithm clearly outperforms a baseline selection method, resulting in a reduction of both the error rate and the required number of variants in the recognition lexicon.
{"title":"Pronunciation variation modeling of non-native proper names by discriminative tree search","authors":"Line Adde, T. Svendsen","doi":"10.1109/ICASSP.2011.5947461","DOIUrl":"https://doi.org/10.1109/ICASSP.2011.5947461","url":null,"abstract":"In this paper, the task of selecting the optimal subset of pronunciation variants from a set of automatically generated candidates is recast as a tree search problem. In this approach, the optimal recognition lexicon corresponds with the optimal path through a search tree. We define a discriminative evaluation function to guide the search algorithm, which is based on estimates of the number of recognition errors before and after a lexicon change. The error rate for a given lexicon is estimated using the Minimum Classification Error framework. Selecting pronunciation candidates by means of this search algorithm clearly outperforms a baseline selection method, resulting in a reduction of both the error rate and the required number of variants in the recognition lexicon.","PeriodicalId":254007,"journal":{"name":"2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129069744","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2011-07-12DOI: 10.1109/ICASSP.2011.5947182
P. Majdak, P. Balázs, W. Kreuzer, M. Dörfler
Exponential sweeps are widely used to measure impulse responses of electro-acoustic systems. Measurements are often contaminated by environmental noise and nonlinear distortions. We propose a method to increase the signal-to-noise ratio (SNR) by denoising the recorded signal in the time-frequency plane. In contrast to state-of-the art denoising methods, no assumption about the spectral characteristics of the noise is required. Numerical simulations demonstrate improvements in the SNR under low-SNR conditions even for measurements contaminated by colored noise.
{"title":"A time-frequency method for increasing the signal-to-noise ratio in system identification with exponential sweeps","authors":"P. Majdak, P. Balázs, W. Kreuzer, M. Dörfler","doi":"10.1109/ICASSP.2011.5947182","DOIUrl":"https://doi.org/10.1109/ICASSP.2011.5947182","url":null,"abstract":"Exponential sweeps are widely used to measure impulse responses of electro-acoustic systems. Measurements are often contaminated by environmental noise and nonlinear distortions. We propose a method to increase the signal-to-noise ratio (SNR) by denoising the recorded signal in the time-frequency plane. In contrast to state-of-the art denoising methods, no assumption about the spectral characteristics of the noise is required. Numerical simulations demonstrate improvements in the SNR under low-SNR conditions even for measurements contaminated by colored noise.","PeriodicalId":254007,"journal":{"name":"2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127903727","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2011-07-12DOI: 10.1109/ICASSP.2011.5946595
Kazuki Nakagami, T. Shiota, T. Nishitani
A simplified shadow removal approach by using interim results of transformed domain GMM foreground segmentation has been developed. The approach is based on the fact that the spatial frequency distribution does not change from the backgrounds in the shadow areas. Due to employing gray level picture processing and to utilizing only low frequency components in the transform domain, the resultant shadow removal approach drastically reduces the amount of processing, compared to conventional shadow removal approaches based on pixel based color component processing.
{"title":"Low complexity shadow removal on foreground segmentation","authors":"Kazuki Nakagami, T. Shiota, T. Nishitani","doi":"10.1109/ICASSP.2011.5946595","DOIUrl":"https://doi.org/10.1109/ICASSP.2011.5946595","url":null,"abstract":"A simplified shadow removal approach by using interim results of transformed domain GMM foreground segmentation has been developed. The approach is based on the fact that the spatial frequency distribution does not change from the backgrounds in the shadow areas. Due to employing gray level picture processing and to utilizing only low frequency components in the transform domain, the resultant shadow removal approach drastically reduces the amount of processing, compared to conventional shadow removal approaches based on pixel based color component processing.","PeriodicalId":254007,"journal":{"name":"2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127487447","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2011-07-12DOI: 10.1109/ICASSP.2011.5946290
Mai Abdelhakim, Lei Zhang, Jian Ren, Tongtong Li
This paper considers cooperative sensing in cognitive networks under Spectrum Sensing Data Falsification attack (SSDF) in which malicious users can intentionally send false sensing information. One effective method to deal with the SSDF attack is the q-out-of-m scheme, where the sensing decision is based on q sensing reports out of m polled nodes. The major limitation with the q-out-of-m scheme is its high computational complexity due to exhaustive search. In this paper, we prove that for a fixed percentage of malicious users, the detection accuracy increases almost exponentially as the network size increases. Motivated by this observation, as well as the linear relationship between the scheme parameters and the network size, we propose a simple but accurate approach that significantly reduces the complexity of the q-out-of-m scheme. The proposed approach can easily be applied to the large scale networks, which can be much more reliable under malicious attacks.
{"title":"Cooperative sensing in cognitive networks under malicious attack","authors":"Mai Abdelhakim, Lei Zhang, Jian Ren, Tongtong Li","doi":"10.1109/ICASSP.2011.5946290","DOIUrl":"https://doi.org/10.1109/ICASSP.2011.5946290","url":null,"abstract":"This paper considers cooperative sensing in cognitive networks under Spectrum Sensing Data Falsification attack (SSDF) in which malicious users can intentionally send false sensing information. One effective method to deal with the SSDF attack is the q-out-of-m scheme, where the sensing decision is based on q sensing reports out of m polled nodes. The major limitation with the q-out-of-m scheme is its high computational complexity due to exhaustive search. In this paper, we prove that for a fixed percentage of malicious users, the detection accuracy increases almost exponentially as the network size increases. Motivated by this observation, as well as the linear relationship between the scheme parameters and the network size, we propose a simple but accurate approach that significantly reduces the complexity of the q-out-of-m scheme. The proposed approach can easily be applied to the large scale networks, which can be much more reliable under malicious attacks.","PeriodicalId":254007,"journal":{"name":"2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122110306","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2011-07-12DOI: 10.1109/ICASSP.2011.5946360
Mojtaba Soltanalian, P. Stoica
In this paper, Perfect Root-of-Unity Codes (PRUCs) with entries in αp = {x ∈ ℂ | xp = 1} where p is a prime are studied. A lower bound on the number of distinct phases in PRUCs over αp is derived. We show that PRUCs of length L ≥ p(p − 1) must use all phases in αp. It is also shown that if there exists a PRUC of length L over αp then p divides L. We derive equations (which we call principal equations) that give possible lengths of a PRUC over αp together with their phase distribution. Using these equations, we prove for example that the length of a 3-phase perfect code must be of the form equation for (h1, h2) ∈ ℤ2 and we also give the exact number of occurences of each element from a3 in the code. Finally, all possible lengths (≤ 100) of PRUCs over α5 and α7 together with their phase distributions are provided.
研究了αp = {x∈| xp = 1}中p为素数的完全统一根码(PRUCs)。推导了PRUCs中不同相数在αp上的下界。我们发现长度L≥p(p−1)的PRUCs必须使用αp中的所有相。还证明了如果存在长度为L / αp的PRUC，则p除L。我们推导出了PRUC的可能长度及其相位分布的方程(我们称之为主方程)。利用这些方程，我们举例证明了三相完美码的长度必须是(h1, h2)∈s2的形式方程，并给出了码中a3中每个元素出现的确切次数。最后给出了α5和α7上所有可能长度(≤100)的PRUCs及其相分布。
{"title":"Perfect Root-Of-Unity Codes with prime-size alphabet","authors":"Mojtaba Soltanalian, P. Stoica","doi":"10.1109/ICASSP.2011.5946360","DOIUrl":"https://doi.org/10.1109/ICASSP.2011.5946360","url":null,"abstract":"In this paper, Perfect Root-of-Unity Codes (PRUCs) with entries in α<inf>p</inf> = {x ∈ ℂ | x<sup>p</sup> = 1} where p is a prime are studied. A lower bound on the number of distinct phases in PRUCs over α<inf>p</inf> is derived. We show that PRUCs of length L ≥ p(p − 1) must use all phases in α<inf>p</inf>. It is also shown that if there exists a PRUC of length L over α<inf>p</inf> then p divides L. We derive equations (which we call principal equations) that give possible lengths of a PRUC over α<inf>p</inf> together with their phase distribution. Using these equations, we prove for example that the length of a 3-phase perfect code must be of the form equation for (h<inf>1</inf>, h<inf>2</inf>) ∈ ℤ<sup>2</sup> and we also give the exact number of occurences of each element from a3 in the code. Finally, all possible lengths (≤ 100) of PRUCs over α<inf>5</inf> and α<inf>7</inf> together with their phase distributions are provided.","PeriodicalId":254007,"journal":{"name":"2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","volume":"669 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115007616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2011-07-12DOI: 10.1109/ICASSP.2011.5946457
Gaurav Sharma, A. Harmanci, D. Mathews
TurboFold, an iterative algorithm for estimating the common secondary structures of multiple RNA homologs, is presented. The algorithm is motivated by and has structure and attributes analogous to the turbo decoding algorithm in communications. Instead of solving the joint problem of aligning and folding multiple RNA sequences, TurboFold uses an iterative process to fold a collection of RNA homologs. Beneficial information from inter-sequence comparisons is incorporated by using feedback from iteration to iteration in the form of pseudo-prior probabilities for base pairing which are incorporated in the computation of base pairing probabilities. As a result Turbo-Fold retains several of the advantages of join alignment and folding while maintaining a per iteration computational complexity comparable to single sequence RNA folding. Experimental evaluation of the algorithm, performed over six ncRNA families, demonstrates that TurboFold achieves high accuracy, offering better performance than available alternatives for estimating RNA base pairing probabilities.
{"title":"Iterative estimation of structures of multiple RNA homologs: Turbofold","authors":"Gaurav Sharma, A. Harmanci, D. Mathews","doi":"10.1109/ICASSP.2011.5946457","DOIUrl":"https://doi.org/10.1109/ICASSP.2011.5946457","url":null,"abstract":"TurboFold, an iterative algorithm for estimating the common secondary structures of multiple RNA homologs, is presented. The algorithm is motivated by and has structure and attributes analogous to the turbo decoding algorithm in communications. Instead of solving the joint problem of aligning and folding multiple RNA sequences, TurboFold uses an iterative process to fold a collection of RNA homologs. Beneficial information from inter-sequence comparisons is incorporated by using feedback from iteration to iteration in the form of pseudo-prior probabilities for base pairing which are incorporated in the computation of base pairing probabilities. As a result Turbo-Fold retains several of the advantages of join alignment and folding while maintaining a per iteration computational complexity comparable to single sequence RNA folding. Experimental evaluation of the algorithm, performed over six ncRNA families, demonstrates that TurboFold achieves high accuracy, offering better performance than available alternatives for estimating RNA base pairing probabilities.","PeriodicalId":254007,"journal":{"name":"2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124316989","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：